No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Artificial Intelligence in Oncology: Current Capabilities, Future Opportunities, and Ethical Considerations
Abstract
The promise of highly personalized oncology care using artificial intelligence (AI) technologies has been forecasted since the emergence of the field. Cumulative advances across the science are bringing this promise to realization, including refinement of machine learning- and deep learning algorithms; expansion in the depth and variety of databases, including multiomics; and the decreased cost of massively parallelized computational power. Examples of successful clinical applications of AI can be found throughout the cancer continuum and in multidisciplinary practice, with computer vision-assisted image analysis in particular having several U.S. Food and Drug Administration-approved uses. Techniques with emerging clinical utility include whole blood multicancer detection from deep sequencing, virtual biopsies, natural language processing to infer health trajectories from medical notes, and advanced clinical decision support systems that combine genomics and clinomics. Substantial issues have delayed broad adoption, with data transparency and interpretability suffering from AI's "black box" mechanism, and intrinsic bias against underrepresented persons limiting the reproducibility of AI models and perpetuating health care disparities. Midfuture projections of AI maturation involve increasing a model's complexity by using multimodal data elements to better approximate an organic system. Far-future positing includes living databases that accumulate all aspects of a person's health into discrete data elements; this will fuel highly convoluted modeling that can tailor treatment selection, dose determination, surveillance modality and schedule, and more. The field of AI has had a historical dichotomy between its proponents and detractors. The successful development of recent applications, and continued investment in prospective validation that defines their impact on multilevel outcomes, has established a momentum of accelerated progress.
... In order to keep the accuracy and generalisability of artificial intelligence algorithms, it is necessary to conduct consistent clinical validation and to use datasets that are robust and diverse. Ethical problems concerning the privacy of data and the interpretability of decisions generated by artificial intelligence are also extremely important and require serious study (Luchini, Pea, Scarpa 2022;Shreve, Khanani, Haddad 2022). As a result of innovations that are anticipated to be focused on drug discovery, the treatment of rare cancers, and the incorporation of real-time patient monitoring through wearable devices (Dlamini, Francies, Hull, Marima 2020), the importance of artificial intelligence in precision oncology is likely to expand in the years to come. ...
... To mitigate this, AI models must be developed using diverse data and undergo regular audits for fairness. Transparency and explainability of AI systems are also vital to building trust among healthcare providers and patients (Shreve, Khanani, Haddad 2022;Castaneda, Nalley, Mannion, et al. 2015). ...
... Regulatory oversight and clear guidelines are important to govern the ethical use of AI in precision oncology (Castaneda, Nalley, Mannion, et al. 2015;Far 2023). This includes establishing protocols for informed consent, ensuring equitable access to AI-driven treatments, and implementing rigorous validation processes to guarantee patient safety and treatment efficacy (Jaremko, Azar, Bromwich, Lum, et al. 2019;Shreve, Khanani, Haddad 2022). Data security measures must be robust to protect against breaches or misuse of sensitive medical information (Jaremko, Azar, Bromwich, Lum, et al. 2019). ...
In precision oncology, the utilization of artificial intelligence (AI) is revolutionising around personalised cancer therapy. After providing an introduction of AI technology and emphasising the importance of precision oncology, this chapter investigates AI applications in oncology. The importance of AI in cancer detection and staging is emphasised, with particular attention on machine learning for precise staging, medical imaging improvements, and AI-assisted pathology. AI examines cancer genomes to identify biomarkers, targetable mutations, and predicts the progression of the disease. A comprehensive description of the link amid AI with genomic profiling is given. More focus is paid on AI-driven decision support systems, combination therapy optimisation, and personalised regimens as well as a deeper examination of individualised treatment planning. The efficacy and integration of AI-based diagnostic tools such as imaging techniques, pathology, histopathology, and liquid biopsies were assessed in this study. This chapter addressed the application of AI in radiation, chemotherapy, surgical decision-making, treatment planning and optimisation. AI-assisted customised dosage and real-time treatment monitoring have enabled significant advancements in adaptive radiation, chemotherapeutic dose, and response tracking. AI's predictive capacity for treatment outcomes and survival rates was investigated, along with its use in immunotherapy and targeted therapies, such as response prediction, CAR-T cell therapy customisation, and assistance in the creation of tailored cancer drugs. This chapter also discusses the technological, ethical, and legal issues surrounding AI in cancer treatment, highlighting the necessity of strong validation procedures and interdisciplinary cooperation. The transformational potential of AI in precision oncology is ultimately highlighted in this chapter, opening the door to more efficient and individualised cancer therapies.
... Estas tecnologias estão em posição de destaque e criam oportunidades adicionais aos profissionais de saúde em termos de capacidade e preparação, resultando em maior integração com a prática clínica (Shreve;Khanani;Haddad, 2022 A presente revisão de escopo foi orientada pela questão de pesquisa: ...
... Estas tecnologias estão em posição de destaque e criam oportunidades adicionais aos profissionais de saúde em termos de capacidade e preparação, resultando em maior integração com a prática clínica (Shreve;Khanani;Haddad, 2022 A presente revisão de escopo foi orientada pela questão de pesquisa: ...
... Estas tecnologias estão em posição de destaque e criam oportunidades adicionais aos profissionais de saúde em termos de capacidade e preparação, resultando em maior integração com a prática clínica (Shreve;Khanani;Haddad, 2022 A presente revisão de escopo foi orientada pela questão de pesquisa: ...
O câncer é a segunda principal causa de morte no mundo. No Brasil, tem sido cada vez mais amplo e intersetorial os investimentos em ações para fortalecer a prevenção, rastreamento, diagnóstico e tratamento, por meio do Sistema Único de Saúde. Frente ao atual arcabouço regulatório brasileiro, a Rede de Atenção à Saúde para o cuidado da pessoa com câncer tem exigido melhoria contínua no gerenciamento dos serviços de saúde que a integram. Diante do potencial da Inteligência Artificial para melhorar as condições de vida e saúde, objetiva-se mapear as evidências científicas sobre algoritmos para gerenciamento de serviços de oncologia hospitalar. Trata-se de uma revisão de escopo conforme recomendação do Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews, para o período de março de 2014 a fevereiro de 2024, nas bases Google Acadêmico; Scielo; Lilacs/BIREME; PubMed e Scopus. Foram identificados 259 artigos e, responderam aos critérios de inclusão apenas cinco estudos. Foram mais prevalentes estudos de revisão e consensos de especialistas. Decision Support System, Machine Learning e Modelagem matemática foram os algoritmos identificados para o gerenciamento dos serviços. Todavia, os estudos foram desenvolvidos no Estados Unidos e na Europa. Fato que denota a necessidade de pesquisas que abordem e induzam o desenvolvimento de algoritmos apropriados ao Sistema Único de Saúde.
... Frequently, ML models operate as black boxes, which makes it difficult to interpret the reasoning behind their predictions [119]. In cancer care, each decision has critical implications, making the development of methods that explain the decision-making process of ML models a requirement essential to build trust and facilitate clinical acceptance. ...
... Introducing ML technologies into clinical practice requires user-friendly tools that integrate seamlessly into workflows and provide clear benefits to healthcare providers and patients [119]. The success of ML-driven systems depends on model interpretability and explainability. ...
... Intellectual property rights and ownership of models, especially those trained on proprietary datasets, present legal challenges. Privacy and data protection regulations, like GDPR, add complexities in data handling, consent management, and cross-border transfers [119]. ...
Machine learning (ML) methods have revolutionized cancer analysis by enhancing the accuracy of diagnosis, prognosis, and treatment strategies. This paper presents an extensive study on the applications of machine learning in cancer analysis, with a focus on three primary areas: a comparative analysis of medical imaging techniques (including X-rays, mammography, ultrasound, CT, MRI, and PET), various AI and ML techniques (such as deep learning, transfer learning, and ensemble learning), and the challenges and limitations associated with utilizing ML in cancer analysis. The study highlights the potential of ML to improve early detection and patient outcomes while also addressing the technical and practical challenges that must be overcome for its effective clinical integration. Finally, the paper discusses future directions and opportunities for advancing ML applications in cancer research.
... This ensures equitable distribution of benefits and optimal performance across subgroups. [52][53][54]57,58 The data should reflect the population, including individuals who are disproportionately affected by the condition or disease. [52][53][54]57,58 The developers of the Paige Prostate Suite used a large number of biopsy slides from institutions around the world. ...
... [52][53][54]57,58 The data should reflect the population, including individuals who are disproportionately affected by the condition or disease. [52][53][54]57,58 The developers of the Paige Prostate Suite used a large number of biopsy slides from institutions around the world. 19 However, the FDA reported that 82.2% of the training data comprised biopsy slides from patients who were white. ...
... 64,65 However, their findings should be validated in Canada. The cost-effectiveness of the suite also requires evaluation, 24,57 with consideration of costs related to changes in productivity. 27 Future prospective trials should consider the impact and diagnostic performance of the suite across race and ethnicity, especially with people who are Black, who are at higher risk of prostate cancer. ...
What Is the Paige Prostate Suite? The Paige Prostate Suite is a set of artificial intelligence (AI) applications that works alongside pathologists reviewing prostate biopsy samples. The suite is not available in Canada as of this writing (June 2024), but international counterparts have authorized it for clinical use. The system requires pathology slides to be digitized for the suite to be able to highlight areas of suspicion for pathologist review. Pathologists can use Paige Prostate Detect as a “second set of eyes” on biopsy slides scanned into the digital system. What Issue Does the Paige Prostate Suite Intend to Address? In Canada, prostate cancer accounted for 20% of new cancers in 2022. It is the most common cancer found in people who have a prostate. Unlike many other cancers, the disease progresses slowly and early diagnosis results in a 5-year survival rate close to 100%. However, health systems are faced with more cancer cases without an increased capacity in pathology. What Is the Potential Impact? Paige Prostate aims to improve pathologists’ ability to detect prostate cancer in less time by allowing pathologists to focus on positive cases. However, the time-saving benefits of the suite require further validation that reflects real world settings and applications. The Paige Prostate Suite has the potential to optimize pathology workflow and prevent delays in diagnosis. Pathologists can use the suite to supplement their review instead of sending biopsy slides to the lab for additional staining or to experts for a second consultation. It does, however, require the added the step of digitizing slides to the workflow. The Paige Prostate Suite aims to improve consistency by helping pathologists grade tumours. What Else Do We Need to Know? Canada is in the early stages of implementing technology to produce digital images of biopsies. To adapt to the uptake of AI, pathology departments will need to adopt new digital workflows and processes. They could leverage this infrastructure for telepathology to improve access in rural and remote areas. Clinicians expect the Paige Prostate Suite to cost more than current practice. Costs will vary depending on the case load, system readiness for AI, and use case for the suite. Cost-effectiveness of the suite requires in-depth investigation that considers the cost-benefit from productivity changes. Upcoming prospective studies will assess the benefits of the Paige Prostate Suite from a clinical utility and cost impact perspective. It is also important to further understand the diagnostic performance and impact of Paige Prostate in the context of race, ethnicity, and other equity considerations, as those in equity-deserving groups may be underrepresented in algorithm development and trials, or have different levels of prostate cancer risk.
... Data should ideally always be up-to-date for the purpose of "AI training." 76 Qualitative and quantitative, highquality data are an important factor in increasing the evidence of AI. To further increase evidence, it is necessary for an AI to be validated by external validation. ...
... 45 Therefore, it is of great importance that training data of AI are representative for all patient groups. 76 The highest possible level of evidence provides patients a high level of safety in treatment. In this context, it is important to know who is legally responsible if an AI in case of potential user mistakes with consequences for the health of patients. ...
... A major goal towards the best possible treatment in oncology is personalized medicine. 76 For this purpose, AI applications allow a maximum high number of data and images of a single patient to be analyzed automatically in a very short time in order to determine the best treatment available 11,78 (Figure 4). This capability is particularly helpful with complex oncological diseases, which to this date are evaluated and treated by MCCs ( Figure 1). ...
Artificial intelligence (AI) promises to be the next revolutionary step in modern society. Yet, its role in all fields of industry and science need to be determined. One very promising field is represented by AI‐based decision‐making tools in clinical oncology leading to more comprehensive, personalized therapy approaches. In this review, the authors provide an overview on all relevant technical applications of AI in oncology, which are required to understand the future challenges and realistic perspectives for decision‐making tools. In recent years, various applications of AI in medicine have been developed focusing on the analysis of radiological and pathological images. AI applications encompass large amounts of complex data supporting clinical decision‐making and reducing errors by objectively quantifying all aspects of the data collected. In clinical oncology, almost all patients receive a treatment recommendation in a multidisciplinary cancer conference at the beginning and during their treatment periods. These highly complex decisions are based on a large amount of information (of the patients and of the various treatment options), which need to be analyzed and correctly classified in a short time. In this review, the authors describe the technical and medical requirements of AI to address these scientific challenges in a multidisciplinary manner. Major challenges in the use of AI in oncology and decision‐making tools are data security, data representation, and explainability of AI‐based outcome predictions, in particular for decision‐making processes in multidisciplinary cancer conferences. Finally, limitations and potential solutions are described and compared for current and future research attempts.
... The complexity of AI algorithms is often associated with lack of transparency, which may result in healthcare professionals feeling uncertain about the reliability of AI applications 146 . Clinicians may hesitate to rely on AI recommendations due to the "black box" nature of many models 147,148 . Explainable AI (XAI) methods are essential for building trust in AI recommendations, helping users to understand the reasoning behind the suggestions, providing transparency, and boosting confidence in the decisions made 149,150 . ...
... Accuracy and reliability. The development of clinical decision support systems is ongoing, and these systems cannot yet be utilized because of the inaccuracy and unreliability of AI predictions 147,148 . Rigorous clinical validation, standardization, and real-world testing are essential before deployment. ...
The confluence of new technologies with artificial intelligence (AI) and machine learning (ML) analytical techniques is rapidly advancing the field of precision oncology, promising to improve diagnostic approaches and therapeutic strategies for patients with cancer. By analyzing multi-dimensional, multiomic, spatial pathology, and radiomic data, these technologies enable a deeper understanding of the intricate molecular pathways, aiding in the identification of critical nodes within the tumor’s biology to optimize treatment selection. The applications of AI/ML in precision oncology are extensive and include the generation of synthetic data, e.g., digital twins, in order to provide the necessary information to design or expedite the conduct of clinical trials. Currently, many operational and technical challenges exist related to data technology, engineering, and storage; algorithm development and structures; quality and quantity of the data and the analytical pipeline; data sharing and generalizability; and the incorporation of these technologies into the current clinical workflow and reimbursement models.
... AI research is advancing in lung cancer screening, potentially improving diagnostic accuracy and Downloaded from https://academic.oup.com/oncolo/advance-article/doi/10.1093/oncolo/oyae169/7713392 by NIH -Health Services Research Library user on 04 October 2024 efficiency, 66 but diagnostic, legal, and ethical challenges may arise from inaccuracies in predictive algorithms. 67 This is particularly concerning for minorities, as machine-learning algorithms are developed based on integrated datasets that underreport underserved communities. 67 Calls to Action for new technologies in lung cancer are shown in Figure 5. ...
... 67 This is particularly concerning for minorities, as machine-learning algorithms are developed based on integrated datasets that underreport underserved communities. 67 Calls to Action for new technologies in lung cancer are shown in Figure 5. ...
Lung cancer, the leading cause of cancer-related deaths globally, remains a pressing health issue despite significant medical advances. The New York Lung Cancer Foundation brought together experts from academia, the pharmaceutical and biotech industries as well as organizational leaders and patient advocates, to thoroughly examine the current state of lung cancer diagnosis, treatment, and research. The goal was to identify areas where our understanding is incomplete and to develop collaborative public health and scientific strategies to generate better patient outcomes, as highlighted in our “Calls to Action.” The consortium prioritized 8 different calls to action. These include (1) develop strategies to cure more patients with early-stage lung cancer, (2) investigate carcinogenesis leading to lung cancers in patients without a history of smoking, (3) harness precision medicine for disease interception and prevention, (4) implement solutions to deliver prevention measures and effective therapies to individuals in under-resourced countries, (5) facilitate collaborations with industry to collect and share data and samples, (6) create and maintain open access to big data repositories, (7) develop new immunotherapeutic agents for lung cancer treatment and prevention, and (8) invest in research in both the academic and community settings. These calls to action provide guidance to representatives from academia, the pharmaceutical and biotech industries, organizational and regulatory leaders, and patient advocates to guide ongoing and planned initiatives.
... These studies are part of intensified efforts to explore the utility of AI in cancer diagnosis, prognosis, prediction and treatments. 13,27,[32][33][34][35] These efforts are progressing along with clinical trials 33,[36][37][38] and an expanding list of AI-based Software as Medical Devices (SaMDs) approved by the Federal Drug Administration (FDA). 39 The largest proportion of these approved devices targets cancer radiology and pathology, 39 representing advances in line with AI pattern recognition and classification capabilities for cancer diagnostics. ...
... The implementation, clinical validation and deployment of On-coGPT would also be a complex undertaking fraught with numerous challenges that are intrinsic to data-driven AI, including explainability and ethical concerns. 36,112 Ongoing efforts to overcome data-related challenges in cancer reasearch 113 and the continuous evolution of the regulatory environment to facilitate a safe, ethical and effective deployment of AI models in the clinic, 39 combined with research advances in AI applications for oncology 32,34 and the exploration of effective paths to their implementations and deployments for patient care 114,115 are providing a dynamic environment for addressing the challenges inherent to datadriven AI and are driving the maturation of the AI-assisted cancer care paradigm. ...
Therapeutic resistance is a major challenge facing the design of effective cancer treatments. Adaptive cancer therapy is in principle the most viable approach to manage cancer’s adaptive dynamics through drug combinations with dose timing and modulation. However, there are numerous open issues facing the clinical success of adaptive therapy. Chief among these issues is the feasibility of real-time predictions of treatment response which represent a bedrock requirement of adaptive therapy. Generative artificial intelligence has the potential to learn prediction models of treatment response from clinical, molecular, and radiomics data about patients and their treatments. The article explores this potential through a proposed integration model of Generative Pre-Trained Transformers (GPTs) in a closed loop with adaptive treatments to predict the trajectories of disease progression. The conceptual model and the challenges facing its realization are discussed in the broader context of artificial intelligence integration in oncology.
... Аналіз останніх досліджень і публікацій. Наукові розвідки щодо перспектив застосування технологій штучного інтелекту у сфері охорони здоров'я здійснювали різні зарубіжні вчені, серед яких варто відзначити дослідження С. Лучіні, А. Пеа та А. Скарпа [11]; Я. Шрів, С. Ханані та Т. Хадада [12]; Ю. Шена та Ф. Шамоута [13]; С. Картер, В. Роджерс, К. Віна, Х. Фрейзер, Б. Річардса та Н. Хуссамі [14]; С. Річардс, Н. Азіза, Ш. Бейл та Д. Біка [15]; С. Беняменса, П. Дунно та Б. Мешко [16]; Д. Шлосека та Дж. Феррета [17]; Е. Ву, К. Ву, Р. Данешджу, Д. Ояна, Д. Хо та Д. Зу [18]. ...
... Науковці з Mayo Clinic (Рочестер, Міннесота, США) Я. Шрів, С. Ханані та Т. Хадад [12] виявили, що використання АІ підвищує результативність планування радіотерапії у процесі контурування пухлин і органів, що, у свою чергу, збільшує швидкість проведення необхідної терапії, та покращує ефективність та безпечність радіаційного використання. Водночас дослідники вважають, що в майбутньому технології АІ можна буде використовувати для виявлення пацієнтів з високим ризиком раку підшлункової залози, використовуючи зображення черевної порожнини та аналізуючи цифрові медичні записи спостереження пацієнтів, що значно підвищить шанси на виявлення хвороби на ранній стадії, а також зможе знизити рівень смертності саме від цього виду онкологічного захворювання. ...
The article is focused on the role and capacity of an AI in combating cancer. It describes achievements and programs of the US National Institute of Cancer in the sphere of machine learning. The article offers the analysis of positive experience of the US National Institute of Cancer in regulatory, organizational and technological cooperation with the United States Department of Health and Human Services and the US Department of Energy in the area of formation and analysis of special registers’medical data with the purpose of effective diagnostics and proper treatment of oncological diseases and development of efficient medicinal drugs. The article presents prospects of national support to the development of new algorithms of AI models in the area of medical assistance to oncological patients. The author describes achievements and challenges to utilization of AI technologies in the US health care system, as well as ethical and legal issues which emerge in this process. It has been proven that absence of a universal Code of regulatory and legal acts of the USA and EU which would regulate algorithms, methods, and machine learning of big data processing in the area of health care leads to the absence of regulated procedures of attributing responsibility in case of harm caused by AI technologies to patients. The author recommends US and EU regulatory authorities to cooperate more actively with international medical research and development institutions (communities) with the aim of drafting relevant legislation and standards. The emphasis is made on a need to provide information protection of rights and freedoms of patients by way of making impossible any capturing of personal medical data irrespective of the form in which it is kept. The conclusion is made that implementation of AI technologies into medical practices is an important factor of government interference aimed at proper management of interaction processes between machines and humans as well as delegation of responsibilities for clinical decisions and potential mistakes.
... Recent improvements concentrate on the incorporation of AI into clinical practice for the four predominant cancer types, involving tasks such as detection, diagnosis, and treatment planning across diverse data modalities [48]. Notwithstanding these encouraging advancements, difficulties persist, such as data transparency, interpretability, and potential biases, which must be resolved for the extensive implementation of AI in oncology [49]. Recent initiatives have highlighted AI's capability in forecasting emergency hospital admissions, and sudden death in clinical trials [50]. ...
Cancer remains a significant health issue, resulting in around 10 million deaths per year, particularly in developing nations. Demographic changes, socio-economic variables, and lifestyle choices are responsible for the rise in cancer cases. Despite the potential to mitigate the adverse effects of cancer by early detection and the implementation of cancer prevention methods, several nations have limited screening facilities. In oncology, the use of artificial intelligence (AI) represents a transformative advancement in cancer diagnosis, prognosis, and treatment. The use of AI in biomarker discovery improves precision medicine by uncovering biomarker signatures that are essential for early detection and treatment of diseases within vast and diverse datasets. Deep learning and machine learning diagnostics are two examples of AI technologies that are changing the way biomarkers are made by finding patterns in large datasets and making new technologies that make it possible to deliver accurate and effective therapies. Existing gaps include data quality, algorithmic transparency, and ethical concerns around privacy, among others. The advancement of biomarker discovery methodologies with AI seeks to transform cancer by improving patient survival rates through enhanced early diagnosis and targeted therapy. This commentary aims to clarify how AI is improving the identification of novel biomarkers for optimal early diagnosis, focused treatment, and improved clinical outcomes, while also addressing certain obstacles and ethical issues related to the application of artificial intelligence in oncology. Data from reputable scientific databases such as PubMed, Scopus, and ScienceDirect were utilized.
... Furthermore, AI-based platforms can synthesize user inputs and provide caregivers with real-time feedback, making caregiving more informed and tailored [26]. Ethical considerations, such as ensuring patient privacy and protecting vulnerable individuals from potential harm, are essential when deploying AI tools in this context [27,28,3]. ...
... In the midterm, AI's evolution is expected to involve increasingly complex models using multimodal data to better simulate biological systems. Looking further ahead, the concept of "living databases" that continuously integrate all aspects of a person's health into comprehensive data sets could drive sophisticated modeling for highly tailored treatments, dose determinations, and surveillance strategies [219]. ...
This comprehensive review explores the integration of Artificial Intelligence (AI) in anticancer drug discovery, highlighting its transformative impact on streamlining the identification, design, and synthesis of novel drug molecules. Leveraging expansive datasets, AI and machine learning technologies enhance the understanding of cancer biology, facilitate target identification, and accelerate the design of molecules with desirable pharmacological properties. Despite promising advancements, challenges persist, including issues related to data quality, model interpretability, and the practical application of AI-generated findings in clinical settings. This review critically examines these challenges, proposes advanced AI models for drug combination predictions, and advocates for collaborative efforts to refine and implement AI methodologies in clinical oncology. The potential of AI to revolutionize anticancer drug discovery is immense, providing a new paradigm that merges precision with efficiency to push the boundaries of therapeutic innovation. Through rigorous validation and interdisciplinary cooperation, AI-driven strategies hold the promise to significantly shorten drug development timelines and improve clinical outcomes, ushering in a new era of personalized medicine in cancer treatment.
... Whole-body PET imaging has great potential for future work, especially the use of artificial intelligence. 27 In line with this, our future work will include segmentation of all disease with lesionby-lesion analysis on W4 and later 18 F-FDG-PET/ CT images in our cohort of patients. With more indepth analysis, we hope to identify specific lesions that do not respond to treatment early in the start of the treatment and offer our patients more personalized treatment. ...
Background
A considerable proportion of metastatic melanoma (mM) patients do not respond to immune checkpoint inhibitors (ICIs). There is a great need to develop noninvasive biomarkers to detect patients, who do not respond to ICIs early during the course of treatment. The aim of this study was to evaluate the role of early [ ¹⁸ F]2fluoro-2-deoxy-D-glucose PET/CT ( ¹⁸ F-FDG PET/CT) at week four (W4) and other possible prognostic biomarkers of survival in mM patients receiving ICIs.
Patients and methods
. In this prospective noninterventional clinical study, mM patients receiving ICIs regularly underwent ¹⁸ F-FDG PET/CT: at baseline, at W4 after ICI initiation, at week sixteen and every 16 weeks thereafter. The tumor response to ICIs at W4 was assessed via modified European Organisation for Research and Treatment of Cancer (EORTC) criteria. Patients with progressive metabolic disease (PMD) were classified into the no clinical benefit group (no-CB), and those with other response types were classified into the clinical benefit group (CB). The primary end point was survival analysis on the basis of the W4 ¹⁸ F-FDG PET/CT response. The secondary endpoints were survival analysis on the basis of LDH, the number of metastatic localizations, and immune-related adverse events (irAEs). Kaplan-Meier analysis and univariate Cox regression analysis were used to assess the impact on survival.
Results
Overall, 71 patients were included. The median follow-up was 37.1 months (952% CI = 30.1–38.0). Three (4%) patients had only baseline scans due to rapid disease progression and death prior to W4 ¹⁸ F-FDG-PET/CT. Fifty-one (72%) patients were classified into the CB group, and 17 (24%) were classified into the no-CB group. There was a statistically significant difference in median overall survival (OS) between the CB group (median OS not reached [NR]; 95% CI = 17.8 months – NR) and the no-CB group (median OS 6.2 months; 95% CI = 4.6 months – NR; p = 0.003). Univariate Cox analysis showed HR of 0.4 (95% CI = 0.18 – 0.72; p = 0.004). median OS was also significantly longer in the group with normal serum LDH levels and the group with irAEs and cutaneous irAEs.
Conclusions
Evaluation of mM patients with early ¹⁸ F-FDG-PET/CT at W4, who were treated with ICIs, could serve as prognostic imaging biomarkers. Other recognized prognostic biomarkers were the serum LDH level and occurrence of cutaneous irAEs.
... www.ijmae.com ethical behavior and responsibility in their technologies (Shreve et al., 2022). Explain Ability: An AI system must be transparent. ...
There are whispers about whether or not Artificial Intelligence (AI) will become more powerful in the field of competition with the human resources of organizations. The main problem for organizations to use AI is the mental ambiguity of their organizational resources about this intelligence in the field of professional ethics. The use of AI in business requires a manager who is familiar with AI issues. Unfortunately, many human resource managers still think AI is a myth, have unscientific and somewhat imaginative expectations, and do not know what transformation AI can bring to their business. This study uses a qualitative meta-analysis method to explore and synthesize existing literature on the role of organizational human ethics based on AI management. Based on the findings of the research, the components of AI management based on professional ethics as a commitment to ethics, ability to explain, fairness, robustness, transparency, privacy protection, international cooperation of use, awareness, use of good data hygiene, use of good data collection, were identified. Controlling users and reducing the algorithmic bias of AI has no place among human intelligence and can only be defined as a helper and not a substitute for humans and the humanity of human resources of organizations. If in planning the development of AI, human and ethical issues are considered together, organizations can hope to realize the dream of ethical and entrepreneurial AI.
... 2. Development of novel theragnostic agents: The development of novel theragnostic agents that combine diagnostic and therapeutic capabilities in a single molecule or nanoparticle could further improve the specificity and sensitivity of therapy monitoring (Thangam et al., 2022;Cheng et al., 2021a). 3. Incorporation of artificial intelligence (AI) and machine learning (ML) algorithms: The use of AI and ML algorithms can enable more accurate and efficient analysis of imaging and biomarker data and facilitate the identification of new biomarkers and treatment targets (Koh et al., 2022;Hou et al., 2022;Shreve et al., 2022). 4. Development of cost-effective theragnostic approaches: The development of cost-effective theragnostic approaches, such as the use of widely available imaging modalities or non-invasive biomarker analysis, could improve the accessibility and affordability of cancer vaccine monitoring for patients (Kemp and Kwon, 2021). ...
Cancer continues to be one of the leading causes of death worldwide, and conventional cancer therapies such as chemotherapy, radiation therapy, and surgery have limitations. RNA therapy and cancer vaccines hold considerable promise as an alternative to conventional therapies for their ability to enable personalized therapy with improved efficacy and reduced side effects. The principal approach of cancer vaccines is to induce a specific immune response against cancer cells. However, a major challenge in cancer immunotherapy is to predict which patients will respond to treatment and to monitor the efficacy of the vaccine during treatment. Theragnostics, an integration of diagnostic and therapeutic capabilities into a single hybrid platform system, has the potential to address these challenges by enabling real-time monitoring of treatment response while allowing endogenously controlled personalized treatment adjustments. In this article, we review the current state-of-the-art in theragnostics for cancer vaccines and RNA therapy, including imaging agents, biomarkers, and other diagnostic tools relevant to cancer, and their application in cancer therapy development and personalization. We also discuss the opportunities and challenges for further development and clinical translation of theragnostics in cancer vaccines.
... 28 Notwithstanding GenAI potential to revolutionize cancer care, 30,52,54 the use of LLMs faces challenges that are typical to ML/AI systems, including the lack of explainability, opacity, ethical concerns, and the need for large training datasets. [55][56][57] In addition, LLMs have an intrinsic risk for hallucinations, 58 which in the context of oncology means that they may yield incorrect or clinically implausible outputs such as nonsensical treatment recommendations. ...
Despite the expanding therapeutic options available to cancer patients, therapeutic resistance, disease recurrence, and metastasis persist as hallmark challenges in the treatment of cancer. The rise to prominence of generative artificial intelligence (GenAI) in many realms of human activities is compelling the consideration of its capabilities as a potential lever to advance the development of effective cancer treatments. This article presents a hypothetical case study on the application of generative pre-trained transformers (GPTs) to the treatment of metastatic prostate cancer (mPC). The case explores the design of GPT-supported adaptive intermittent therapy for mPC. Testosterone and prostate-specific antigen (PSA) are assumed to be repeatedly monitored while treatment may involve a combination of androgen deprivation therapy (ADT), androgen receptor-signalling inhibitors (ARSI), chemotherapy, and radiotherapy. The analysis covers various questions relevant to the configuration, training, and inferencing of GPTs for the case of mPC treatment with a particular attention to risk mitigation regarding the hallucination problem and its implications to clinical integration of GenAI technologies. The case study provides elements of an actionable pathway to the realization of GenAI-assisted adaptive treatment of metastatic prostate cancer. As such, the study is expected to help facilitate the design of clinical trials of GenAI-supported cancer treatments.
... Further, the review also discusses AI's role in trying to forecast outcomes for the patients as another area of development that has occurred. Through clinical information, features of the tumour, type of mutation the patient has or earlier response to a certain treatment, AI algorithms (ML , DL, predictive analysis etc) can identify how the patient will react to a particular form of treatment or recurrence risk (31). ...
... Addressing these challenges necessitates the development of more robust algorithms and the incorporation of large-scale, diverse datasets to ensure broader applicability (Huang et al., 2014). A growing body of literature has systematically reviewed the application of AI/ML techniques in brain tumor detection, shedding light on their potential and limitations (Shreve et al., 2022). Anjum et al., (2021) conducted an extensive review of transfer learning applications, highlighting its ability to overcome data scarcity by leveraging pre-trained models. ...
1,126 articles were initially identified, with rigorous screening and quality assessments narrowing the final selection to 102 high-quality studies. The review highlights the superior diagnostic accuracy and efficiency achieved by AI models, particularly Convolutional Neural Networks (CNNs), which consistently report accuracy rates exceeding 90%. Hybrid and ensemble models further enhance diagnostic robustness, addressing challenges related to complex tumor types and heterogeneous datasets. Data-related issues, such as scarcity and imbalance, remain critical barriers, with studies emphasizing the effectiveness of synthetic data generation and augmentation techniques in improving model generalizability. Explainable AI (XAI) frameworks have been identified as pivotal for fostering clinician trust, offering interpretability and transparency that facilitate integration into clinical workflows. Real-time diagnostic systems demonstrate the potential for AI to streamline operations and enable timely clinical decisions, particularly in resource-constrained settings. Despite these advancements, challenges such as algorithmic bias, data diversity, and infrastructural limitations persist. This review underscores the transformative role of AI/ML in brain tumor diagnostics, providing actionable insights to advance research and clinical adoption, ultimately improving patient outcomes and healthcare efficiency.
... On the other hand, ethical and economic implications, especially in low-resource settings, must also be addressed-whether cancer is prematurely classified as CUP is also a matter of how much resources can be invested in diagnosis. Future research should focus on developing cost-effective, computationally efficient models that maintain high accuracy, promoting equitable access to precision medicine for CUP patients globally [30,31]. Lastly, clinical validation is essential. ...
... A primary concern is data privacy and security, especially regarding the collection of personal data without consent, as seen in the use of RFID technology [42]. In the healthcare sector, for instance, the application of AI raises concerns about algorithmic bias that could impact patient diagnosis outcomes, highlighting the importance of ethical guidelines to protect patient rights [43,44]. Another challenge includes the social impact of new technologies, such as deepfake, on public trust in democracy [45]. ...
Background. The emergence of digital tools, such as data analytics and automation, enhances decision-making, aligns with sustainability, and supports strategic goals, especially in the context of achieving Sustainable Development Goals (SDGs) Research Purpose. The study aims to provide a comprehensive analysis of the developmental trends in management accounting by systematically reviewing recent literature. This review focuses on identifying key trends, existing gaps, and emerging themes to establish a framework for future research in this evolving field? ResearchMethod. Using a systematic literature review (SLR) approach, the research includes journal articles, conference proceedings, and reports published over the past decade. Articles were selected based on relevance to management accounting and sorted by topic, findings, and theoretical contributions.? Findings. Key trends in management accounting include the integration of digital technology (e.g., AI, blockchain) for efficiency, accuracy, and improved decision-making, alongside a strong focus on environmental sustainability. The study also found significant interest in sustainability management practices, particularly regarding their role in meeting SDGs and enhancing corporate performance ?Conclusion. Management accounting is increasingly shaped by digitalization and sustainability imperatives. Technologies like AI and blockchain are reshaping data accuracy and efficiency, while sustainability practices encourage responsible business practices.
... As with great power comes bigger economic impact, personalized healthcare requires large sums of investments and some of the underrepresented or minority groups may have limited access to such novel technologies (14). This coincides with Eroom's law, which describes the ever-slowing rate of drug discovery and applicability with increasing costs associated with it (15). ...
Precision medicine is revolutionizing the world in combating different disease modalities, including cancer. The concept of personalized treatments is not new, but modeling it into a reality has faced various limitations. The last decade has seen significant improvements in incorporating several novel tools, scientific innovations and governmental support in precision oncology. However, the socio-economic factors and risk-benefit analyses are important considerations. This mini review includes a summary of some commendable milestones, which are not just a series of successes, but also a cautious outlook to the challenges and practical implications of the advancing techno-medical era.
... Telehealth is similarly cost-effective compared to center-based rehabilitation. Data from clinical trials show reduced work disability and rehospitalization rates, [61] which can serve as an essential argument for integrating into routine practice [62]. Despite improved outcomes in quality of life and survivorship, insurance coverage and reimbursement for exercise rehabilitation remains exclusive to specific comorbid conditions and not cancer diagnosis per se. ...
Purpose of Review
A critical evaluation of contemporary literature regarding the role of big data, artificial intelligence, and digital technologies in precision cardio-oncology care and survivorship, emphasizing innovative and groundbreaking endeavors.
Recent Findings
Artificial intelligence (AI) algorithm models can automate the risk assessment process and augment current subjective clinical decision tools. AI, particularly machine learning (ML), can identify medically significant patterns in large data sets. Machine learning in cardio-oncology care has great potential in screening, diagnosis, monitoring, and managing cancer therapy-related cardiovascular complications. To this end, large-scale imaging data and clinical information are being leveraged in training efficient AI algorithms that may lead to effective clinical tools for caring for this vulnerable population. Telemedicine may benefit cardio-oncology patients by enhancing healthcare delivery through lowering costs, improving quality, and personalizing care. Similarly, the utilization of wearable biosensors and mobile health technology for remote monitoring holds the potential to improve cardio-oncology outcomes through early intervention and deeper clinical insight. Investigations are ongoing regarding the application of digital health tools such as telemedicine and remote monitoring devices in enhancing the functional status and recovery of cancer patients, particularly those with limited access to centralized services, by increasing physical activity levels and providing access to rehabilitation services.
Summary
In recent years, advances in cancer survival have increased the prevalence of patients experiencing cancer therapy-related cardiovascular complications. Traditional cardio-oncology risk categorization largely relies on basic clinical features and physician assessment, necessitating advancements in machine learning to create objective prediction models using diverse data sources. Healthcare disparities may be perpetuated through AI algorithms in digital health technologies. In turn, this may have a detrimental effect on minority populations by limiting resource allocation. Several AI-powered innovative health tools could be leveraged to bridge the digital divide and improve access to equitable care.
... 9 Most clinical notes, including those from oncologists, mid-level professionals and nurses, present unique machine curation challenges due to stylistic variations as they comprise narrative text notes, typed or dictated, that form semi-structured and unstructured content. 10,11 The variability in expression, form, and content within these notes underscores the need for advanced data processing methods. Additional clinical data related to genomics and molecular diagnostics further complicate patient qualification for clinical trials. ...
Clinical trial research in oncology relies heavily on clinical documentation within the electronic medical record (EMR) to ascertain patient eligibility in clinical trials based on inclusion and exclusion criteria. The structured data elements within the EMR serve as the primary information source for defining patient cohorts, with clinical cancer stage and performance status being two pivotal criteria determining trial eligibility. The challenge arises from the inconsistent availability of clinical stage and performance status data within the structured fields of the EMR despite their consistent presence in clinical notes. Additionally, there is a deficiency of standardization of this data that exists in the unstructured field. Hence, due to lack of structured data and standardization of said data, there are limitations in developing artificial intelligence (AI) models. To increase the comprehensiveness of clinical records, a clinical research team at a community oncology practice was consulted to identify requirements and extract essential clinical features from de-identified data. The methods outlined in this paper focused on eliminating false positives to allow future development of Large Language Models (LLM) using the outputted structured fields which resulted in an increase in patient record completeness with high accuracy. The accuracy ranged from 97.5-97.75% for the models that were developed. Out of the 60,000+ patients, the numerical staging, TNM (tumor, node, metastasis) staging, and Karnofsky performance score models added a structured field for 29.62%, 21.01%, and 40.64% patients respectively. Additionally, a semi-supervised NLP algorithm was applied on the performance status algorithm which achieved a mean absolute error (MAE) of 1.57. This work demonstrates the use case of natural language processing (NLP) in optimizing the clinical research enrollment process by providing an efficient and accurate method to detect key clinical values in unstructured patient data. Similar methodology with more advanced algorithms such as LLM can be employed to detect additional patient elements such as molecular biomarkers, imaging reports, postoperative surgical outcomes (i.e., clear margins etc.) and patient treatment outcomes using the extracted structured fields.
... The utilization of big data and AI, including natural language processing and medical record analysis, holds tremendous potential in uncovering insights that can utilize large datasets for predictive capability (Bakken et al., 2020;Keim-Malpass & Kausch, 2023). Research priorities around these potential applications are diverse, encompassing algorithm development for optimal cancer treatment, symptom management, quality care aligned with patient goals, optimal workflow, and equitable care through the thoughtful integration of available social demographics of health to predict patient vulnerability (Mema & McGinty, 2020;Shreve et al., 2022). In addition, generative AI, or "chatbots," derived from large language models, will not only influence clinical care as it becomes more precise and sophisticated but also open exciting new directions for the numerous oncology nursing research trajectories based on developing accurate and trustworthy patient counseling, education, and support (Iannantuono et al., 2023;Kolla & Parikh, 2024). ...
Problem statement:
To define the Oncology Nursing Society Research Agenda for 2024-2027.
Design:
An iterative, multiple data sources consolidation through the Research Agenda Project Team.
Data sources:
Previous research priorities, literature review, stakeholder survey, and research priorities from other cancer care organizations and funding agencies.
Findings:
10 evergreen statements articulated foundational values for oncology nurse scientists, and 5 topics emerged as research priorities for the upcoming three years: Advance patient-centric, precision symptom science; provide evidence for safe and effective cancer care delivery models and support of the oncology nursing workforce; describe the impact of the environment on cancer care outcomes; integrate patient navigation into cancer care across the trajectory; and advance the use of innovative methodologies in oncology nursing research.
Implications for nursing:
The Oncology Nursing Society Research Agenda is an effective resource for directing the organization's research vision. This foundational document directs funding awards and requests, mentorship, and policy initiatives.
... Challenges in AI model development, such as data standardisation, bias, the need for large datasets and limited validation, can be overcome with concurrent investments. In oncology, early AI tools in EHRs show promise, visualising patient health like a digital photograph (43). However, ethical considerations in AI cancer research, including data privacy, consent, and fairness, demand careful attention. ...
The rapid evolution in artificial intelligence (AI) technology has led to
the development of a range of softwares and tools within cancer diagnostics and therapeutics. This ranges
from imaging, risk and outcome prediction, drug development and triage, and holds great promise in
revolutionising cancer medicine by increasing efficiency, reducing waiting times and facilitating the provision
of personalised medicine. However, various obstacles impede widespread integration of AI tools into clinical
settings such as the black box nature of AI, data shift and ethical implications relating to the storage of
personal data and how it is used. The aim of this narrative review is to inform the reader on current uses of
AI in cancer care, and explore avenues for potential future uses and developments.
... However, despite the exciting progress being made, there are challenges to the application of AI in cancer research. The need to ensure the interpretability of algorithms, address ethical issues and resolve biases in data are critical aspects that require ongoing attention [14]. This review examines AI applications to genomic data in cancer research, focusing on five key areas: ...
This review focuses on the intersection of artificial intelligence and genomic data in cancer research. It explores the types of genomic data used in the literature, the methodologies of machine learning and deep learning, recent applications, and the challenges associated with this field. Through an analysis of 47,586 articles and addressing seven research questions, the study reveals significant growth in this area over the past years. While there has been remarkable progress, ongoing attention is needed to address ethical considerations, interpretability of algorithms, and potential data biases, to ensure the reliable and responsible use of these advanced technologies. Overall, this paper provides a comprehensive overview of the current research landscape, offering insights into both the potential and challenges of AI in genomic data research.
... The results of the experiments show that accuracy was achieved for SVM by (93%), Decision Trees (94%), Naïve Base (93%), and CNN (90%). Jacob T. et al. [20] proposed a description for present AI capabilities, upcoming possibilities, and moral issues in oncology a CNN is used to detect tumors automatically. ...
The use of deep learning for identifying defects in medical images has rapidly emerged as a significant area of interest across various medical diagnostic applications. The automated recognition of Posterior Fossa Tumors (PFT) in Magnetic Resonance Imaging (MRI) plays a vital role, as it furnishes essential data about irregular tissue, essential for treatment planning. Human examination has traditionally been the standard approach for identifying defects in brain MRI. This technique is unsuitable for a massive quantity of data. Therefore, automated PFT detection techniques are being established to minimize radiologist's time. In this paper, the posterior fossa tumor is detected and classified in brain MRI using Convolutional Neural Network (CNN) algorithms, and the model result and accuracy obtained from each algorithm are explained. A dataset collection made up of 3,00,000 images with an average of 500 patients from the Children's Cancer Hospital Egypt (CCHE) was used. The CNN algorithms investigated to classify the PFT were VGG19, VGG16, and ResNet50. Moreover, explanations for the behavior of networks were investigated using three different techniques: LIME, SHAP, and ICE. Overall, the results showed that the best model was VGG16 compared with other CNN-used models with accuracy rate values of 95.33%, 93.25%, and 87.4%, respectively.
... This can result in a lack of treatments for diseases that affect marginalized populations or for rare diseases. Bias can also influence the selection of molecular targets, leading to the exclusion of promising drug candidates that may be effective for specific patient populations [53,54]. ...
... Artificial intelligence (AI) is gaining ground in the current precision oncology era [101]. In this field, AI could be useful for clinicians to identify patients at higher risk of DDI: implementing this tool will be important to optimize therapeutic choices, limiting toxicities and avoidable interactions. ...
Introduction:
The therapeutic scenario of metastatic hormone-sensitive prostate cancer (mHSPC) has dramatically changed in recent years, with the approval of new-generation Androgen Receptor Signaling Inhibitors (ARSIs), in combination with the androgen deprivation therapy (ADT), which was the previous standard of care. Despite showing a similar clinical efficacy, ARSIs, all of which are administered orally, are different in terms of pharmacokinetic and drug-drug interactions (DDIs).
Areas covered:
This review covers the main pharmacokinetic characteristics of ARSIs that have been approved for the first-line therapy of mHSPC patients, underlying the differences among these molecules and focusing on the known or possible interactions with other drugs. Full-text articles and abstracts were searched in PubMed.
Expert opinion:
Since prostate cancer occurs mainly in older age, comorbidities and the consequent polypharmacy increase the DDI risk in mHSPC patients who are candidates for ARSI. Waiting for new therapeutic options, in the absence of direct comparisons, pharmacokinetic knowledge is essential to guide clinicians in prescribing ARSI in this setting.
Introduction
Artificial intelligence (AI) has significant potential to improve health outcomes in oncology. However, as AI utility increases, it is imperative to ensure that these models do not systematize racial and ethnic bias and further perpetuate disparities in health. This scoping review evaluates the transparency of demographic data reporting and diversity of participants included in published clinical studies utilizing AI in oncology.
Methods
We utilized PubMed to search for peer‐reviewed research articles published between 2016 and 2021 with the query type “(“deep learning” or “machine learning” or “neural network” or “artificial intelligence”) and (“neoplas” or “tumor”).” We included clinical trials and original research studies and excluded reviews and meta‐analyses. Oncology‐related studies that described data sets used in training or validation of the AI models were eligible. Data regarding public reporting of patient demographics were collected, including age, sex at birth, and race. We used descriptive statistics to analyze these data across studies.
Results
Out of 220 total studies, 118 were eligible and 47 (40%) had at least one described training or validation data set publicly available. 69 studies (58%) reported age data for patients included in training or validation sets, 60 studies (51%) reported sex, and six studies (5%) reported race. Of the studies that reported race, a range of 70.7%–93.4% of individuals were White. Only three studies reported racial demographic data with greater than two categories (i.e. “White” vs. “non‐White” or “White” vs. “Black”).
Conclusions
We found that a minority of studies (5%) analyzed reported racial and ethnic demographic data. Furthermore, studies that did report racial demographic data had few non‐White patients. Increased transparency regarding reporting of demographics and greater representation in data sets is essential to ensure fair and unbiased clinical integration of AI in oncology.
By providing cutting-edge therapeutic interventions, improving accessibility, and creating immersive healing experiences, generative artificial intelligence (AI) and virtual reality (VR) are transforming mental health care. This study investigates how these new technologies affect mental health, looking at how they might enhance wellbeing while resolving ethical issues and inherent difficulties. By offering individualized interventions, cognitive behavioral therapy (CBT), and real-time emotional support, generative AI-powered chatbots and virtual assistants lower obstacles to mental health care. By establishing safe, immersive settings that promote gradual desensitization, virtual reality exposure therapy has shown promise in the treatment of phobias, anxiety disorders, and post-traumatic stress disorder (PTSD). Notwithstanding these benefits, issues with algorithmic bias, data privacy, and an excessive dependence on technology pose serious problems. To guarantee patient safety, the ethical ramifications of AI-generated mental health advice-specifically, its accuracy and dependability-need thorough confirmation. Guidelines for ethical use are necessary because VR's immersive nature can also result in dissociation, addiction, or unexpected psychological impacts. To ensure fair access and efficacy, technologists, psychologists, and legislators must work together to create standardized guidelines for integrating AI and VR into clinical practice. The implications of AI-driven mental health interventions for marginalized groups-who frequently face inequities in access to conventional care-are also examined in this research. Generative AI models' versatility makes it possible to create therapeutic applications that are inclusive of all languages and cultures, filling in gaps in mental health care around the globe. To solve issues with AI bias, false information, and responsibility in automated mental health solutions, legislative and ethical frameworks must change. The use of AI and VR in self-guided therapy, crisis intervention, and preventive care is growing as these technologies continue to transform the field of mental health. Research on these technologies' long-term effects on social connections, human emotional intelligence, and psychological resilience is still lacking, despite the fact that they have the potential to improve patient participation and lessen the workload for mental health practitioners. To optimize the advantages of generative AI and VR for mental health, this study emphasizes the need to strike a balance between technology innovation and human-centric ethical issues.
Background
Artificial intelligence (AI) is a boon to the human race that offers transformative potential in the medical care system, revolutionizing human well-being. Over the past five decades, AI has evolved significantly in deep learning and machine learning (ML). AI subfields work together to provide intelligence for various applications. ML is a self-learning system that can improve its performance through training experiences. Utilizing artificial neural networks mimics human brain functions, while computer vision involves computers extracting information from images or videos. The application of AI is deployed across diverse medical fields, including cardiology, dermatology, ophthalmology, and oncology, enhancing diagnostic procedures and treatment outcomes.
Objective
This review aims to explore current trends of AI in healthcare, evaluate its impact across different medical fields, and identify future prospects for AI-driven innovations in personalized medicine and beyond.
Method
A comprehensive literature analysis was undertaken using prominent databases such as “PubMed,” “Scopus,” and “Google Scholar.”
Results
The review found that AI has significantly impacted multiple areas of healthcare. In diagnostics, AI applications have improved accuracy and efficiency, particularly in fields such as cardiology and oncology. Overall, while AI holds promise for revolutionizing healthcare, its success will depend on addressing the challenges and continuing to advance both technology and implementation practices.
Il capitolo "L’intelligenza artificiale a supporto dell’invecchiamento attivo" esplora il ruolo dell'IA nel migliorare la qualità della vita degli anziani in un contesto di crescente invecchiamento della popolazione globale. Attraverso tecnologie come sensori intelligenti, algoritmi di apprendimento automatico e robot sociali, l’IA consente il monitoraggio della salute, interventi personalizzati e la promozione di uno stile di vita attivo e sicuro. Vengono analizzati approcci di apprendimento supervisionato e non supervisionato, esempi concreti di progetti innovativi e le prospettive future, evidenziando i benefici e le sfide etiche legate alla privacy e alla sicurezza dei dati.
Artificial intelligence is rapidly changing our world at an exponential rate and its transformative power has extensively reached important sectors like healthcare. In the fight against cancer, AI proved to be a novel and powerful tool, offering new hope for prevention and early detection. In this review, we will comprehensively explore the medical applications of AI, including early cancer detection through pathological and imaging analysis, risk stratification, patient triage, and the development of personalized prevention approaches. However, despite the successful impact AI has contributed to, we will also discuss the myriad of challenges that we have faced so far toward optimal AI implementation. There are problems when it comes to the best way in which we can use AI systemically. Having the correct data that can be understood easily must remain one of the most significant concerns in all its uses including sharing information. Another challenge that exists is how to interpret AI models because they are too complicated for people to follow through examples used in their developments which may affect trust, especially among medical professionals. Other considerations like data privacy, algorithm bias, and equitable access to AI tools have also arisen. Finally, we will evaluate possible future directions for this promising field that highlight AI’s capacity to transform preventative cancer care.
The future of medicine promises an exciting transformation driven by unprecedented technological advancements, personalized healthcare practices, and collaborative interdisciplinary research. This abstract explores key trends and innovations shaping the landscape of medicine in the coming years. The application of artificial intelligence and machine learning to improve diagnoses and therapy, the creation of targeted therapies based on genetic profiling, and the integration of wearable technologies and telemedicine for remote patient monitoring are among the developments. Furthermore, the rise of precision medicine and a greater emphasis on preventive treatment are predicted to improve patient outcomes and overall healthcare delivery. The integration of artificial intelligence (AI) and machine learning algorithms into diagnostics has brought in a new era of precision medicine, enabling faster and more precise disease diagnosis and prognosis. The rise of genomic medicine has opened up the potential for tailored treatments, with breakthroughs in gene editing technologies and the development of targeted therapies. CRISPR-based interventions hold promise for curing genetic conditions, while developments in regenerative medicine and 3D bioprinting open the door to organ transplantation. Privacy, data security, and the responsible use of emerging technologies are among the issues that ethical considerations and regulatory frameworks are addressing as they change quickly to keep up with the developments. This chapter only introduces key trends and innovations shaping the landscape of medicine in the coming years.
Artificial Intelligence (AI) has the potential to revolutionize medical diagnostics by offering new opportunities for accuracy, efficiency, and accessibility in healthcare. This article examines the benefits of implementing AI in diagnostics, such as enhanced diagnostic precision, faster clinical decision-making, cost reduction, and increased access to healthcare. It also discusses the challenges associated with AI implementation, including ethical, legal, and technical issues. The future of AI in medicine may bring further technological advancements and personalized therapy, but it also requires addressing regulatory and ethical concerns.
A BSTRACT
Artificial intelligence (AI) is revolutionizing health care by addressing some of the important concerns, the health-care organizations face daily. All partners in the health system must understand AI technologies and how they might improve the effectiveness and accessibility of AI-based health services, leading to value-based care. Effective and proper use of AI in health care is the primary emphasis of this narrative review article, which also helps readers grasp the basic ideas underlying AI. Despite the fact that AI is still in its infancy in other sectors of health care, it has made tremendous strides in a variety of specializations, including radiodiagnosis and imaging, surgery (robotic-assisted procedures), oncology, especially radiation oncology, anesthesia, and pathology. However, ethical concerns about utilizing AI in health care may delay its widespread usage.
Cancer is one of the main causes of death for people worldwide. Breast, lung, colon, brain and lymphoma are some of the most common types of cancer. Successful treatment can significantly increase the chances of survival. Enhancing the probability of a successful cancer treatment requires initial identification and treatment. In this paper a model is proposed using denset121 pretrained model with modified dense net block and softmax function as output layer. There are two subgroups of the total number of diseases: task 1 and task 2. Task1 include breast, kidney, cervical, leukemia while task2 include lung, oral, lymphoma, brain.A person suffering from the disease of task 1 may also suffer from a disease belonging to task 2. This model is examined using a dataset with multiple cancers, which is publicly available on Kaggle. The suggested method performs with an accuracy of 99.31% for task 1 as well as 97.02% for task 2, respectively, when analyzed alongside the most recent techniques.
The integration of biomaterials in oncology has revolutionized cancer treatment strategies, offering promising avenues for personalized and targeted therapies. One of the key challenges in cancer treatment is the delivery of therapeutic agents specifically to tumor sites while minimizing systemic toxicity. Biomaterial-based drug delivery systems have emerged as a promising solution, offering controlled release and targeted delivery of chemotherapeutic agents, immunotherapeutics, and gene therapies. By encapsulating drugs within biocompatible carriers such as nanoparticles, liposomes, and hydrogels, biomaterials enable sustained release kinetics, enhancing therapeutic efficacy while reducing side effects. Biomaterials play a crucial role in tissue engineering approaches for cancer therapy. Scaffold-based constructs provide a supportive microenvironment for the growth and differentiation of cells, facilitating the regeneration of damaged tissues and organs. In oncology, biomaterial scaffolds can be engineered to mimic the extracellular matrix of tumors, enabling the culture of patient-derived cancer cells for drug screening, disease modeling, and personalized medicine applications. In the realm of cancer diagnostics, biomaterials offer innovative solutions for early detection and monitoring of disease. Nanotechnology-based biosensors and imaging probes leverage the unique physicochemical properties of biomaterials to detect biomarkers associated with cancer progression. By enhancing the sensitivity and specificity of diagnostic assays, biomaterials enable early-stage detection of tumors, guiding treatment decisions and improving patient outcomes. The field of biomaterials in oncology is poised for continued growth and innovation.
Der vorliegende Beitrag beschäftigt sich mit den unterschiedlichen Anwendungsbereichen der Künstlichen Intelligenz in der Onkologie. Der Schwerpunkt liegt hierbei auf den Herausforderungen in dem technisch noch vergleichsweisen neuen Anwendungsbereich der medizinischen Therapieempfehlung. In den vergangenen Jahren wurden bereits diverse Anwendungen Künstlicher Intelligenz in der Medizin entwickelt, die hauptsächlich auf der Analyse von Bilddateien (z. B. in der Radiologie oder Pathologie) beruhen. Das primäre Ziel hierbei ist die Unterstützung der Ärzte in der alltäglichen Auswertung einer großen Anzahl an Befunde, um dessen Qualität zu erhöhen und Fehler zu reduzieren. Im Zentrum der Behandlung von Patienten mit Tumorerkrankungen steht die interdisziplinäre Therapieempfehlung. Diese hoch-komplexe Entscheidung beruht auf einer großen Anzahl an Informationen (sowohl der Patienten als auch der potenziellen Therapieoptionen), die in kürzester Zeit analysiert und richtig eingeordnet werden müssen. In diesem Beitrag beschreiben wir die technischen und medizinischen und Anforderungen einer Künstlichen Intelligenz, um diese noch wenig erforschte Herausforderung interdisziplinär anzugehen und potenzielle Lösungsansätze zu entwickeln.
Artificial intelligence (AI) has advanced quickly and shown promise in revolutionizing biomedical cancer research by providing cutting-edge methods for diagnosing cancer. AI has strong logical independent learning ability, which can enhance the process of improving cancer-related work. By determining the risk factors, predictive-AI models can calculate an individual's probability of developing cancer. AI tools like machine learning have the potential to significantly improve the anticancer drug research paradigm as it now exists. Through the use of sophisticated image analysis techniques, AI has completely changed the field of cancer diagnosis by enabling early detection and accurate characterization. AI-driven algorithms improve the precision of medical imaging in radiodiagnosis, making them a vital tool for the medical team in the identification and evaluation of cancer. With its potential to provide more personalized and comprehensive care, AI in oncology has enormous promise for improving patient outcomes in the near future. Artificial Intelligence (AI) is genuinely changing our lives and creating significant new prospects in the oncology sector, which is essentially enhancing the care of cancer patients. The need for more uniformity, effectiveness, and efficiency in cancer care across the several oncology workflow domains is in line with the potential of AI in this field. AI has been used in the field of cancer research for approximately 20 years, but it becomes more effective in recent few years because of a significant amount of molecular-level tumor data from cancer patients is now available due to the rapid increase in both volume and variety of data sets. Expert-level performance has been achieved and promising advances have been made in the field of cancer research. Numerous companies and institutions of higher learning are using AI to detect, diagnose, and treat cancer. The first AI-enabled machine for cancer radiology (IB Neuro, Version 1.0, by Imaging Biometrics, LLC.) was approved by the FDA on May 15, 2008. A recent GE Healthcare and MIT (Massachusetts Institute of Technology) Review found that 79% of physicians believe AI technologies have speed up operations, allowing medical professionals to offer better services and patient-centered care. The key potential benefits of utilizing AI-enabled technology and supporting clinical decisions in oncology include: Automated cancer diagnosis Enhanced prediction capabilities Real-time data updates Personalized attention Increased efficiency, improved results, and reduced costs Cancer is a complex and multifaceted disorder with thousands of genetic and epigenetic variations, especially in how they grow and divide. These technologies can improve treatment accuracy and personalization by analyzing complex data. This may improve patient outcomes and reduce adverse effects. AI and ML(Machine Learning) prediction models can detect high-risk individuals and enable early cancer treatment. In this presentation we are trying to give our overview about involvement of AI in Oncology and covered the necessary field to the best of our knowledge. Machine learning is a specialized field within AI that refers to a group of algorithms designed to automatically learn and improve from experience. In other words, machine learning is an AI subset that focuses on developing algorithms capable of learning from data and refining their performance over time. Deep Learning is a subfield of "Machine Learning" that employs neural network-based models to imitate the human brain's capacity to analyze huge amounts of complicated data in areas such as language processing, drug discovery, and image recognition. The term Artificial Intelligence (AI) describes the intelligence exhibited by human-made machines basically it is a set of computer-coded programs or algorithms that use data analysis and pre-programmed instructions to make predictions and decisions. It is an interdisciplinary field that encompasses computer science, mathematics and biological algorithms. AI is originally formalized in 1950s, It is a collection of "self-learning" iterative methods that find patterns in data that can change and frequently do so more quickly over time. Following decades of rapid progress, Artificial Intelligence (AI) is now a catch-all term for a variety of technologies, including deep learning, machine learning, and artificial neural networks. AI's ability to transform early cancer diagnosis and address capacity issues through automation has been demonstrated in relation to health care data. It is possible to conclude from this presentation that AI will enable us to examine complicated data from a variety of processes, such as genomic, metabolomic, radiomic, and clinical text data. The technological advantages of this period have allowed AI-ML to grow. Further medical and technical developments are required to help with improved therapy because of the rising prevalence and mortality of cancer. By recognising the risk variables, predictive AI models are able to determine an individual's chance of developing a certain kind of cancer. AI can help us live a happier and healthier life and beat the sickness known as cancer if we can reduce the limitations and challenges. We are grateful for the support and motivation provided by our mentor, Dr. Subhalakshmi Ghosh and Dr. Aminul Islam. We also thank the Ramakrishna Mission Vivekananda Centenary College and Academy of Biodiversity conservation for giving us the opportunity to present a poster on this great platform. The future of AI in cancer research holds tremendous promise for improving patient outcomes through advanced and individualized care as well as AI-driven models are going to be incorporated into every aspect of healthcare. In the upcoming years, oncology AI applications will happen through data intelligence, better tumor understanding, more precise treatment options and improved decision-making processes. It also can developed as a risk assessment tool integrated into smart phone applications which will be able to instantly evaluate their risk of developing cancer. AI will be marked as a revolutionary tool in the field of healthcare specially in oncology.
AI's integration and adoption in the sector have evolved to be a game-changer through operational revolutionisation regarding accessibility to advanced diagnosis and treatments, reduced waiting times, and cost savings. This chapter explores the strategic efficacy of AI in the context of medical tourism. Using the term “strategic efficacy,” the authors encompass the concept of efficiency and effectiveness of AI in achieving a strategic outcome in medical tourism. The authors' purviews are that is important to ensure that an AI strategy in medical tourism not only looks good on paper but also continues to produce high success for the global practice. In this chapter, the authors discuss AI's emergence in the medical tourism industry, the strategic efficacy of AI in medical tourism, the categories of AI-system devices used in medical tourism, and the AI-system devices. Also discussed are AI systems applications to some major diseases in medical tourism.
Objective:
To perform a scoping review of imaging-based machine-learning models to predict clinical outcomes and identify biomarkers in patients with PDAC.
Summary of background data:
Patients with PDAC could benefit from better selection for systemic and surgical therapy. Imaging-based machine-learning models may improve treatment selection.
Methods:
A scoping review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses-scoping review guidelines in the PubMed and Embase databases (inception-October 2020). The review protocol was prospectively registered (open science framework registration: m4cyx). Included were studies on imaging-based machine-learning models for predicting clinical outcomes and identifying biomarkers for PDAC. The primary outcome was model performance. An area under the curve (AUC) of ≥0.75, or a P-value of ≤0.05, was considered adequate model performance. Methodological study quality was assessed using the modified radiomics quality score.
Results:
After screening 1619 studies, 25 studies with 2305 patients fulfilled the eligibility criteria. All but 1 study was published in 2019 and 2020. Overall, 23/25 studies created models using radiomics features, 1 study quantified vascular invasion on computed tomography, and one used histopathological data. Nine models predicted clinical outcomes with AUC measures of 0.78-0.95, and C-indices of 0.65-0.76. Seventeen models identified biomarkers with AUC measures of 0.68-0.95. Adequate model performance was reported in 23/25 studies. The methodological quality of the included studies was suboptimal, with a median modified radiomics quality score score of 7/36.
Conclusions:
The use of imaging-based machine-learning models to predict clinical outcomes and identify biomarkers in patients with PDAC is increasingly rapidly. Although these models mostly have good performance scores, their methodological quality should be improved.
Deep learning is a subdiscipline of artificial intelligence that uses a machine learning technique called artificial neural networks to extract patterns and make predictions from large data sets. The increasing adoption of deep learning across healthcare domains together with the availability of highly characterised cancer datasets has accelerated research into the utility of deep learning in the analysis of the complex biology of cancer. While early results are promising, this is a rapidly evolving field with new knowledge emerging in both cancer biology and deep learning. In this review, we provide an overview of emerging deep learning techniques and how they are being applied to oncology. We focus on the deep learning applications for omics data types, including genomic, methylation and transcriptomic data, as well as histopathology-based genomic inference, and provide perspectives on how the different data types can be integrated to develop decision support tools. We provide specific examples of how deep learning may be applied in cancer diagnosis, prognosis and treatment management. We also assess the current limitations and challenges for the application of deep learning in precision oncology, including the lack of phenotypically rich data and the need for more explainable deep learning models. Finally, we conclude with a discussion of how current obstacles can be overcome to enable future clinical utilisation of deep learning.
Background
There exists scant evidence on the optimal approaches to integrating patient-reported outcomes (PROs) in clinical practice. This study gathered oncology practitioners’ experiences with implementing PROs in cancer care.Methods
Between December 2019 and June 2020, we surveyed practitioners who reported spending > 5% of their time providing clinical care to cancer patients. Respondents completed an online survey describing their experiences with and barriers to using PROs in clinical settings.ResultsIn total, 362 practitioners (physicians 38.7%, nurses 46.7%, allied health professionals 14.6%) completed the survey, representing 41 countries (Asia–Pacific 42.5%, North America 30.1%, Europe 24.0%, others 3.3%). One quarter (25.4%) identified themselves as “high frequency users” who conducted PRO assessments on > 80% of their patients. Practitioners commonly used PROs to facilitate communication (60.2%) and monitor treatment responses (52.6%). The most commonly reported implementation barriers were a lack of technological support (70.4%) and absence of a robust workflow to integrate PROs in clinical care (61.5%). Compared to practitioners from high-income countries, more practitioners in low-middle income countries reported not having access to a local PRO expert (P < .0001) and difficulty in identifying the appropriate PRO domains (P = .006). Compared with nurses and allied health professionals, physicians were more likely to perceive disruptions in clinical care during PRO collection (P = .001) as an implementation barrier.Conclusions
Only a quarter of the surveyed practitioners reported capturing PROs in routine clinical practice. The implementation barriers to PRO use varied across respondents in different professions and levels of socioeconomic resources. Our findings can be applied to guide planning and implementation of PRO collection in cancer care.
Several sources of bias can affect the performance of machine learning systems used in medicine and potentially impact clinical care. Here, we discuss solutions to mitigate bias across the different development steps of machine learning-based systems for medical applications. Vokinger et al. discuss potential sources of bias in machine learning systems used in medicine. The authors propose solutions to mitigate bias across the different stages of model development, from data collection and preparation to model evaluation and application.
Introduction
Artificial intelligence-based tools can significantly impact detection and segmentation of brain metastases for stereotactic radiosurgery (SRS). VBrain is a deep learning algorithm, recently FDA-cleared, to assist in brain tumor contouring. In this study, we aimed to further validate this tool in patients treated with SRS for brain metastases at Stanford Cancer Center.
Methods
We included randomly selected patients with brain metastases treated with SRS from 2008 to 2020. Computed tomography (CT) and axial T1-weighted post-contrast magnetic resonance (MR) image data were extracted for each patient and uploaded to VBrain. Subsequent analyses compared the output contours from VBrain with the physician-defined contours used for SRS. A brain metastasis was considered “detected” when the VBrain “predicted” contours overlapped with the corresponding physician contours (“ground-truth” contours). We evaluated performance against ground-truth contours using the following metrics: lesion-wise Dice similarity coefficient (DSC), lesion-wise average Hausdorff distance (AVD), false positive count (FP), and lesion-wise sensitivity (%).
Results
We analyzed 60 patients with 321 intact brain metastases treated over 70 SRS courses. Resection cavities were excluded from the analysis. The median (range) tumor size was 132 mm3 (7 to 24,765). Input CT scan slice thickness was 1.250 mm, and median (range) pixel resolution was 0.547 mm (0.457 to 0.977). Input MR scan median (range) slice thickness was 1.000 mm (0.940 to 2.000), and median (range) pixel resolution was 0.469 mm (0.469 to 1.094). In assessing VBrain performance, we found mean lesion-wise DSC to be 0.70, mean lesion-wise AVD to be 9.40% of lesion size (0.805 mm), mean FP to be 0.657 tumors per case, and lesion-wise sensitivity to be 84.5%.
Conclusion
Retrospective analysis of our brain metastases cohort using a deep learning algorithm yielded promising results. Integration of VBrain into the clinical workflow can provide further clinical and research insights.
Due to the upfront role of magnetic resonance imaging (MRI) for prostate cancer (PCa) diagnosis, a multitude of artificial intelligence (AI) applications have been suggested to aid in the diagnosis and detection of PCa. In this review, we provide an overview of the current field, including studies between 2018 and February 2021, describing AI algorithms for (1) lesion classification and (2) lesion detection for PCa. Our evaluation of 59 included studies showed that most research has been conducted for the task of PCa lesion classification (66%) followed by PCa lesion detection (34%). Studies showed large heterogeneity in cohort sizes, ranging between 18 to 499 patients (median = 162) combined with different approaches for performance validation. Furthermore, 85% of the studies reported on the stand-alone diagnostic accuracy, whereas 15% demonstrated the impact of AI on diagnostic thinking efficacy, indicating limited proof for the clinical utility of PCa AI applications. In order to introduce AI within the clinical workflow of PCa assessment, robustness and generalizability of AI applications need to be further validated utilizing external validation and clinical workflow experiments.
In recent years, the field of artificial intelligence (AI) in oncology has grown exponentially. AI solutions have been developed to tackle a variety of cancer-related challenges. Medical institutions, hospital systems, and technology companies are developing AI tools aimed at supporting clinical decision making, increasing access to cancer care, and improving clinical efficiency while delivering safe, high-value oncology care. AI in oncology has demonstrated accurate technical performance in image analysis, predictive analytics, and precision oncology delivery. Yet, adoption of AI tools is not widespread, and the impact of AI on patient outcomes remains uncertain. Major barriers for AI implementation in oncology include biased and heterogeneous data, data management and collection burdens, a lack of standardized research reporting, insufficient clinical validation, workflow and user-design challenges, outdated regulatory and legal frameworks, and dynamic knowledge and data. Concrete actions that major stakeholders can take to overcome barriers to AI implementation in oncology include training and educating the oncology workforce in AI; standardizing data, model validation methods, and legal and safety regulations; funding and conducting future research; and developing, studying, and deploying AI tools through multidisciplinary collaboration.
Objective
We performed a meta‐analysis to systematically summarize previous studies on the accuracy of AI in the detection of esophageal cancer and neoplasms in endoscopic images, so as to provide scientific evidence for the effectiveness of AI‐assisted diagnostic models.
Methods
We searched the databases of PubMed, Embase and Cochrane for studies on AI‐assisted esophageal cancer and neoplasms in endoscopic images. We used a bivariate mixed‐effects binary regression model to calculate the pooled diagnostic efficacy of AI. Subgroup analyses and meta‐regression were performed to explore the sources of heterogeneity. We also compared the effectiveness of AI with that of endoscopists.
Result
A total of 16 studies were included in our meta‐analysis. The pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio and AUC were 0.94 (95%CI: 0.92‐0.96), 0.85 (95%CI: 0.73‐0.92), 6.397 (95%CI: 3.380‐12.106), 0.065 (95%CI: 0.041‐0.102), 98.881 (95%CI: 39.446‐247.865) and 0.97 (95%CI: 0.95‐0.98), respectively. AI‐based models showed a better performance than endoscopists in terms of the pooled sensitivity (0.94; 95%CI: 0.84‐0.98 vs. 0.82; 95%CI: 0.77‐0.86, p < 0.01).
Conclusion
The use of AI was shown to achieve a high accuracy in the detection of early esophageal cancer. However, most of the studies were based on retrospective review of selected images, further validation in prospective trials is expected.
In Low- and Middle- Income Countries (LMICs), machine learning (ML) and artificial intelligence (AI) offer attractive solutions to address the shortage of health care resources and improve the capacity of the local health care infrastructure. However, AI and ML should also be used cautiously, due to potential issues of fairness and algorithmic bias that may arise if not applied properly. Furthermore, populations in LMICs can be particularly vulnerable to bias and fairness in AI algorithms, due to a lack of technical capacity, existing social bias against minority groups, and a lack of legal protections. In order to address the need for better guidance within the context of global health, we describe three basic criteria (Appropriateness, Fairness, and Bias) that can be used to help evaluate the use of machine learning and AI systems: 1) APPROPRIATENESS is the process of deciding how the algorithm should be used in the local context, and properly matching the machine learning model to the target population; 2) BIAS is a systematic tendency in a model to favor one demographic group vs another, which can be mitigated but can lead to unfairness; and 3) FAIRNESS involves examining the impact on various demographic groups and choosing one of several mathematical definitions of group fairness that will adequately satisfy the desired set of legal, cultural, and ethical requirements. Finally, we illustrate how these principles can be applied using a case study of machine learning applied to the diagnosis and screening of pulmonary disease in Pune, India. We hope that these methods and principles can help guide researchers and organizations working in global health who are considering the use of machine learning and artificial intelligence.
Background
Screening patients for eligibility for clinical trials is labor intensive. It requires abstraction of data elements from multiple components of the longitudinal health record and matching them to inclusion and exclusion criteria for each trial. Artificial intelligence (AI) systems have been developed to improve the efficiency and accuracy of this process.
Objective
This study aims to evaluate the ability of an AI clinical decision support system (CDSS) to identify eligible patients for a set of clinical trials.
Methods
This study included the deidentified data from a cohort of patients with breast cancer seen at the medical oncology clinic of an academic medical center between May and July 2017 and assessed patient eligibility for 4 breast cancer clinical trials. CDSS eligibility screening performance was validated against manual screening. Accuracy, sensitivity, specificity, positive predictive value, and negative predictive value for eligibility determinations were calculated. Disagreements between manual screeners and the CDSS were examined to identify sources of discrepancies. Interrater reliability between manual reviewers was analyzed using Cohen (pairwise) and Fleiss (three-way) κ, and the significance of differences was determined by Wilcoxon signed-rank test.
ResultsIn total, 318 patients with breast cancer were included. Interrater reliability for manual screening ranged from 0.60-0.77, indicating substantial agreement. The overall accuracy of breast cancer trial eligibility determinations by the CDSS was 87.6%. CDSS sensitivity was 81.1% and specificity was 89%.
Conclusions
The AI CDSS in this study demonstrated accuracy, sensitivity, and specificity of greater than 80% in determining the eligibility of patients for breast cancer clinical trials. CDSSs can accurately exclude ineligible patients for clinical trials and offer the potential to increase screening efficiency and accuracy. Additional research is needed to explore whether increased efficiency in screening and trial matching translates to improvements in trial enrollment, accruals, feasibility assessments, and cost.
Early cancer detection should lead to an overall stage shift, less-intensive treatments and better patient outcomes. Current recommended screening programmes are limited to a handful of individual cancers. A multi-cancer early detection test that simultaneously detects and localises multiple cancers could reduce the morbidity and mortality associated with cancer.
Background
Pathologic complete response (pCR) to neoadjuvant chemotherapy (NAC) is strongly associated with favorable outcome. We examined the utility of serial circulating tumor DNA (ctDNA) testing for predicting pCR and risk of metastatic recurrence.
Patients and methods
Cell-free DNA (cfDNA) was isolated from 291 plasma samples of 84 high-risk early breast cancer patients treated in the neoadjuvant I-SPY 2 TRIAL with standard NAC alone or combined with MK-2206 (AKT inhibitor) treatment. Blood was collected at pretreatment (T0), 3 weeks after initiation of paclitaxel (T1), between paclitaxel and anthracycline regimens (T2), or prior to surgery (T3). A personalized ctDNA test was designed to detect up to 16 patient-specific mutations (from whole exome sequencing of pretreatment tumor) in cfDNA by ultra-deep sequencing. The median follow-up time for survival analysis was 4.8 years.
Results
At T0, 61 of 84 (73%) patients were ctDNA-positive, which decreased over time (T1-35%; T2-14%; T3-9%). Patients who remained ctDNA-positive at T1 were significantly more likely to have residual disease after NAC (83% non-pCR) compared to those who cleared ctDNA (52% non-pCR; OR 4.33, P=0.012). After NAC, all patients who achieved pCR were ctDNA-negative (n=17, 100%). For those who did not achieve pCR (n=43), ctDNA-positive patients (14%) had significantly increased risk of metastatic recurrence (HR 10.4; 95% CI, 2.3–46.6); interestingly, patients who did not achieve pCR but were ctDNA-negative (86%) had excellent outcome, similar to those who achieved pCR (HR 1.4; 95% CI, 0.15–13.5).
Conclusions
Lack of ctDNA clearance was a significant predictor of poor response and metastatic recurrence, while clearance was associated with improved survival even in patients who did not achieve pCR. Personalized monitoring of ctDNA during NAC of high-risk early breast cancer may aid in real-time assessment of treatment response and help fine-tune pCR as a surrogate endpoint of survival.
Several biomarkers such as tumor mutation burden (TMB), neoantigen load (NAL), programmed cell-death receptor 1 ligand (PD-L1) expression, and lactate dehydrogenase (LDH) have been developed for predicting response to immune checkpoint inhibitors (ICIs) in melanoma. However, some limitations including the undefined cut-off value, poor uniformity of test platform, and weak reliability of prediction have restricted the broad application in clinical practice. In order to identify a clinically actionable biomarker and explore an effective strategy for prediction, we developed a genetic mutation model named as immunotherapy score (ITS) for predicting response to ICIs therapy in melanoma, based on whole-exome sequencing data from previous studies. We observed that patients with high ITS had better durable clinical benefit and survival outcomes than patients with low ITS in three independent cohorts, as well as in the meta-cohort. Notably, the prediction capability of ITS was more robust than that of TMB. Remarkably, ITS was not only an independent predictor of ICIs therapy, but also combined with TMB or LDH to better predict response to ICIs than any single biomarker. Moreover, patients with high ITS harbored the immunotherapy-sensitive characteristics including high TMB and NAL, ultraviolet light damage, impaired DNA damage repair pathway, arrested cell cycle signaling, and frequent mutations in NF1 and SERPINB3/4. Overall, these findings deserve prospective investigation in the future and may help guide clinical decisions on ICIs therapy for patients with melanoma.
Background:
Ovarian cancer survival rates have not changed in the last 20 years. The majority of cases are High-grade serous ovarian carcinomas (HGSOCs), which are typically diagnosed at an advanced stage with multiple metastatic lesions. Taking biopsies of all sites of disease is infeasible, which challenges the implementation of stratification tools based on molecular profiling.
Main body:
In this review, we describe how these challenges might be overcome by integrating quantitative features extracted from medical imaging with the analysis of paired genomic profiles, a combined approach called radiogenomics, to generate virtual biopsies. Radiomic studies have been used to model different imaging phenotypes, and some radiomic signatures have been associated with paired molecular profiles to monitor spatiotemporal changes in the heterogeneity of tumours. We describe different strategies to integrate radiogenomic information in a global and local manner, the latter by targeted sampling of tumour habitats, defined as regions with distinct radiomic phenotypes.
Conclusion:
Linking radiomics and biological correlates in a targeted manner could potentially improve the clinical management of ovarian cancer. Radiogenomic signatures could be used to monitor tumours during the course of therapy, offering additional information for clinical decision making. In summary, radiogenomics may pave the way to virtual biopsies and treatment monitoring tools for integrative tumour analysis.
Glioblastoma is the most common malignant brain parenchymal tumor yet remains challenging to treat. The current standard of care—resection and chemoradiation—is limited in part due to the genetic heterogeneity of glioblastoma. Previous studies have identified several tumor genetic biomarkers that are frequently present in glioblastoma and can alter clinical management. Currently, genetic biomarker status is confirmed with tissue sampling, which is costly and only available after tumor resection or biopsy. The purpose of this study was to evaluate a fully automated artificial intelligence approach for predicting the status of several common glioblastoma genetic biomarkers on preoperative MRI. We retrospectively analyzed multisequence preoperative brain MRI from 199 adult patients with glioblastoma who subsequently underwent tumor resection and genetic testing. Radiomics features extracted from fully automated deep learning-based tumor segmentations were used to predict nine common glioblastoma genetic biomarkers with random forest regression. The proposed fully automated method was useful for predicting IDH mutations (sensitivity = 0.93, specificity = 0.88), ATRX mutations (sensitivity = 0.94, specificity = 0.92), chromosome 7/10 aneuploidies (sensitivity = 0.90, specificity = 0.88), and CDKN2 family mutations (sensitivity = 0.76, specificity = 0.86).
Objective
To systematically examine the design, reporting standards, risk of bias, and claims of studies comparing the performance of diagnostic deep learning algorithms for medical imaging with that of expert clinicians.
Design
Systematic review.
Data sources
Medline, Embase, Cochrane Central Register of Controlled Trials, and the World Health Organization trial registry from 2010 to June 2019.
Eligibility criteria for selecting studies
Randomised trial registrations and non-randomised studies comparing the performance of a deep learning algorithm in medical imaging with a contemporary group of one or more expert clinicians. Medical imaging has seen a growing interest in deep learning research. The main distinguishing feature of convolutional neural networks (CNNs) in deep learning is that when CNNs are fed with raw data, they develop their own representations needed for pattern recognition. The algorithm learns for itself the features of an image that are important for classification rather than being told by humans which features to use. The selected studies aimed to use medical imaging for predicting absolute risk of existing disease or classification into diagnostic groups (eg, disease or non-disease). For example, raw chest radiographs tagged with a label such as pneumothorax or no pneumothorax and the CNN learning which pixel patterns suggest pneumothorax.
Review methods
Adherence to reporting standards was assessed by using CONSORT (consolidated standards of reporting trials) for randomised studies and TRIPOD (transparent reporting of a multivariable prediction model for individual prognosis or diagnosis) for non-randomised studies. Risk of bias was assessed by using the Cochrane risk of bias tool for randomised studies and PROBAST (prediction model risk of bias assessment tool) for non-randomised studies.
Results
Only 10 records were found for deep learning randomised clinical trials, two of which have been published (with low risk of bias, except for lack of blinding, and high adherence to reporting standards) and eight are ongoing. Of 81 non-randomised clinical trials identified, only nine were prospective and just six were tested in a real world clinical setting. The median number of experts in the comparator group was only four (interquartile range 2-9). Full access to all datasets and code was severely limited (unavailable in 95% and 93% of studies, respectively). The overall risk of bias was high in 58 of 81 studies and adherence to reporting standards was suboptimal (<50% adherence for 12 of 29 TRIPOD items). 61 of 81 studies stated in their abstract that performance of artificial intelligence was at least comparable to (or better than) that of clinicians. Only 31 of 81 studies (38%) stated that further prospective studies or trials were required.
Conclusions
Few prospective deep learning studies and randomised trials exist in medical imaging. Most non-randomised trials are not prospective, are at high risk of bias, and deviate from existing reporting standards. Data and code availability are lacking in most studies, and human comparator groups are often small. Future studies should diminish risk of bias, enhance real world clinical relevance, improve reporting and transparency, and appropriately temper conclusions.
Study registration
PROSPERO CRD42019123605.
This study aimed to validate the long-term prognostic value of a new clinical-genomic model, Distant Genetic Model-Clinical Variable Model 6 (DGM-CM6), developed in Asia as a prognostic panel for all subtypes of breast cancer. We included 752 operable stage I–III breast cancer patients representing all subtypes treated from 2005 to 2014 as the validation cohort. The median follow-up was 95.8 months. The low- and high-risk patients classified by DGM-CM6 (RI-DR) had significant differences in 10-year distant recurrence-free interval (DRFI) (94.1% vs. 85.0%, P < 0.0001) and relapse-free survival (RFS) (90.0% vs. 80.5%, P = 0.0003). External validation using EMTAB-365 dataset showed similar observation (P < 0.0001). DGM-CM6 was an independent prognostic factor by multivariate analysis with hazard ratios of 3.1 (1.6–6.0) for RFS (P = 0.0009) and 3.8 (1.6–9.0) for DRFI (P = 0.0028). Comparing the C-index of DGM-CM6 and PAM50-ROR scores, the former performed better than the latter in predicting long-term DRFI and RFS, especially in N0, ER/PR-positive, and HER2-negative patients.
Background
Estimation of the risk of malignancy in pulmonary nodules detected by CT is central in clinical management. The use of artificial intelligence (AI) offers an opportunity to improve risk prediction. Here we compare the performance of an AI algorithm, the lung cancer prediction convolutional neural network (LCP-CNN), with that of the Brock University model, recommended in UK guidelines.
Methods
A dataset of incidentally detected pulmonary nodules measuring 5–15 mm was collected retrospectively from three UK hospitals for use in a validation study. Ground truth diagnosis for each nodule was based on histology (required for any cancer), resolution, stability or (for pulmonary lymph nodes only) expert opinion. There were 1397 nodules in 1187 patients, of which 234 nodules in 229 (19.3%) patients were cancer. Model discrimination and performance statistics at predefined score thresholds were compared between the Brock model and the LCP-CNN.
Results
The area under the curve for LCP-CNN was 89.6% (95% CI 87.6 to 91.5), compared with 86.8% (95% CI 84.3 to 89.1) for the Brock model (p≤0.005). Using the LCP-CNN, we found that 24.5% of nodules scored below the lowest cancer nodule score, compared with 10.9% using the Brock score. Using the predefined thresholds, we found that the LCP-CNN gave one false negative (0.4% of cancers), whereas the Brock model gave six (2.5%), while specificity statistics were similar between the two models.
Conclusion
The LCP-CNN score has better discrimination and allows a larger proportion of benign nodules to be identified without missing cancers than the Brock model. This has the potential to substantially reduce the proportion of surveillance CT scans required and thus save significant resources.
Artificial intelligence (AI) has contributed substantially to the resolution of a variety of biomedical problems, including cancer, over the last decade. Deep learning, a subfield of AI that is highly flexible and supports automatic feature extraction, is increasingly being applied in various areas of both basic and clinical cancer research. In this review, we describe numerous recent examples of the application of AI in oncology-including cases in which deep learning has efficiently solved problems that were previously thought to be unsolvable-and we address obstacles that must be overcome before such application can become more widespread. We also highlight resources and data sets that can help harness the power of AI for cancer research. The development of innovative approaches to and applications of AI will yield important insights in oncology in the coming decade.
Artificial intelligence-powered medical technologies are rapidly evolving into applicable solutions for clinical practice. Deep learning algorithms can deal with increasing amounts of data provided by wearables, smartphones, and other mobile monitoring sensors in different areas of medicine. Currently, only very specific settings in clinical practice benefit from the application of artificial intelligence, such as the detection of atrial fibrillation, epilepsy seizures, and hypoglycemia, or the diagnosis of disease based on histopathological examination or medical imaging. The implementation of augmented medicine is long-awaited by patients because it allows for a greater autonomy and a more personalized treatment, however, it is met with resistance from physicians which were not prepared for such an evolution of clinical practice. This phenomenon also creates the need to validate these modern tools with traditional clinical trials, debate the educational upgrade of the medical curriculum in light of digital medicine as well as ethical consideration of the ongoing connected monitoring. The aim of this paper is to discuss recent scientific literature and provide a perspective on the benefits, future opportunities and risks of established artificial intelligence applications in clinical practice on physicians, healthcare institutions, medical education, and bioethics.
Tree-based machine learning models such as random forests, decision trees and gradient boosted trees are popular nonlinear predictive models, yet comparatively little attention has been paid to explaining their predictions. Here we improve the interpretability of tree-based models through three main contributions. (1) A polynomial time algorithm to compute optimal explanations based on game theory. (2) A new type of explanation that directly measures local feature interaction effects. (3) A new set of tools for understanding global model structure based on combining many local explanations of each prediction. We apply these tools to three medical machine learning problems and show how combining many high-quality local explanations allows us to represent global structure while retaining local faithfulness to the original model. These tools enable us to (1) identify high-magnitude but low-frequency nonlinear mortality risk factors in the US population, (2) highlight distinct population subgroups with shared risk characteristics, (3) identify nonlinear interaction effects among risk factors for chronic kidney disease and (4) monitor a machine learning model deployed in a hospital by identifying which features are degrading the model’s performance over time. Given the popularity of tree-based machine learning models, these improvements to their interpretability have implications across a broad set of domains. Tree-based machine learning models are widely used in domains such as healthcare, finance and public services. The authors present an explanation method for trees that enables the computation of optimal local explanations for individual predictions, and demonstrate their method on three medical datasets.
A large body of literature is available on wound healing in humans. Nonetheless, a standardized ex vivo wound model without disruption of the dermal compartment has not been put forward with compelling justification. Here, we present a novel wound model based on application of negative pressure and its effects for epidermal regeneration and immune cell behaviour. Importantly, the basement membrane remained intact after blister roof removal and keratinocytes were absent in the wounded area. Upon six days of culture, the wound was covered with one to three-cell thick K14+Ki67+ keratinocyte layers, indicating that proliferation and migration were involved in wound closure. After eight to twelve days, a multi-layered epidermis was formed expressing epidermal differentiation markers (K10, filaggrin, DSG-1, CDSN). Investigations about immune cell-specific manners revealed more T cells in the blister roof epidermis compared to normal epidermis. We identified several cell populations in blister roof epidermis and suction blister fluid that are absent in normal epidermis which correlated with their decrease in the dermis, indicating a dermal efflux upon negative pressure. Together, our model recapitulates the main features of epithelial wound regeneration, and can be applied for testing wound healing therapies and investigating underlying mechanisms.
Screening mammography aims to identify breast cancer at earlier stages of the disease, when treatment can be more successful¹. Despite the existence of screening programmes worldwide, the interpretation of mammograms is affected by high rates of false positives and false negatives². Here we present an artificial intelligence (AI) system that is capable of surpassing human experts in breast cancer prediction. To assess its performance in the clinical setting, we curated a large representative dataset from the UK and a large enriched dataset from the USA. We show an absolute reduction of 5.7% and 1.2% (USA and UK) in false positives and 9.4% and 2.7% in false negatives. We provide evidence of the ability of the system to generalize from the UK to the USA. In an independent study of six radiologists, the AI system outperformed all of the human readers: the area under the receiver operating characteristic curve (AUC-ROC) for the AI system was greater than the AUC-ROC for the average radiologist by an absolute margin of 11.5%. We ran a simulation in which the AI system participated in the double-reading process that is used in the UK, and found that the AI system maintained non-inferior performance and reduced the workload of the second reader by 88%. This robust assessment of the AI system paves the way for clinical trials to improve the accuracy and efficiency of breast cancer screening.
Breast cancer care is a leading area for development of artificial intelligence (AI), with applications including screening and diagnosis, risk calculation, prognostication and clinical decision-support, management planning, and precision medicine. We review the ethical, legal and social implications of these developments. We consider the values encoded in algorithms, the need to evaluate outcomes, and issues of bias and transferability, data ownership, confidentiality and consent, and legal, moral and professional responsibility. We consider potential effects for patients, including on trust in healthcare, and provide some social science explanations for the apparent rush to implement AI solutions. We conclude by anticipating future directions for AI in breast cancer care. Stakeholders in healthcare AI should acknowledge that their enterprise is an ethical, legal and social challenge, not just a technical challenge. Taking these challenges seriously will require broad engagement, imposition of conditions on implementation, and pre-emptive systems of oversight to ensure that development does not run ahead of evaluation and deliberation. Once artificial intelligence becomes institutionalised, it may be difficult to reverse: a proactive role for government, regulators and professional groups will help ensure introduction in robust research contexts, and the development of a sound evidence base regarding real-world effectiveness. Detailed public discussion is required to consider what kind of AI is acceptable rather than simply accepting what is offered, thus optimising outcomes for health systems, professionals, society and those receiving care.
Background:
Artificial intelligence (AI) is increasingly being used in healthcare. Here, AI-based chatbot systems can act as automated conversational agents, capable of promoting health, providing education, and potentially prompting behaviour change. Exploring the motivation to use health chatbots is required to predict uptake; however, few studies to date have explored their acceptability. This research aimed to explore participants' willingness to engage with AI-led health chatbots.
Methods:
The study incorporated semi-structured interviews (N-29) which informed the development of an online survey (N-216) advertised via social media. Interviews were recorded, transcribed verbatim and analysed thematically. A survey of 24 items explored demographic and attitudinal variables, including acceptability and perceived utility. The quantitative data were analysed using binary regressions with a single categorical predictor.
Results:
Three broad themes: 'Understanding of chatbots', 'AI hesitancy' and 'Motivations for health chatbots' were identified, outlining concerns about accuracy, cyber-security, and the inability of AI-led services to empathise. The survey showed moderate acceptability (67%), correlated negatively with perceived poorer IT skills OR = 0.32 [CI95%:0.13-0.78] and dislike for talking to computers OR = 0.77 [CI95%:0.60-0.99] as well as positively correlated with perceived utility OR = 5.10 [CI95%:3.08-8.43], positive attitude OR = 2.71 [CI95%:1.77-4.16] and perceived trustworthiness OR = 1.92 [CI95%:1.13-3.25].
Conclusion:
Most internet users would be receptive to using health chatbots, although hesitancy regarding this technology is likely to compromise engagement. Intervention designers focusing on AI-led health chatbots need to employ user-centred and theory-based approaches addressing patients' concerns and optimising user experience in order to achieve the best uptake and utilisation. Patients' perspectives, motivation and capabilities need to be taken into account when developing and assessing the effectiveness of health chatbots.
Objective
Clinical prediction models (CPMs) constructed based on artificial intelligence have been proven to have positive impacts on clinical activities. However, the deterioration of CPM performance over time has rarely been studied. This paper proposes a model updating method to solve the calibration drift issue caused by data drift.
Materials and methods
This paper proposes a novel model updating method based on lifelong machine learning (LML). The effectiveness of the proposed method is verified in four tumor datasets, and a comprehensive comparison with other model updating methods is performed.
Results
Changes in data distributions cause model performances to drift. The four compared model updating methods have different effects in terms of improving the discrimination and calibration abilities of the tested models. The LML method proposed in this study improves model performance better than or equivalent to the other methods. The proposed method achieved a mean AUC of 0.8249, 0.8780, 0.8261, and 0.8489, a mean AUPRC of 0.7782, 0.9730, 0.4655, and 0.5728, a mean F1 of 0.6866, 0.9552, 0.2985, and 0.3585, and a mean estimated calibration index (ECI) of 0.0320, 0.0338, 0.0101, and 0.0115 using colorectal, lung, breast and prostate cancer datasets.
Discussion
The LML framework simultaneously monitors model performance and the distribution of disease risk characteristics, enabling it to effectively address the performance degradation caused by gradual and sudden data drifts and provide reasonable explanations for the causes of performance degradation.
Conclusion
Monitoring model performance and the underlying data distribution can promote model life cycle iteration with “development-deployment-maintenance-monitoring” as the core, which, in turn, ensures that the model can provide accurate predictions, guides the model update process and explains the causes of model performance changes.
Cancers other than breast, colorectal, cervical, and lung do not have guideline-recommended screening. New multi-cancer early detection (MCED) tests-using a single blood sample-have been developed based on circulating cell-free DNA (cfDNA) or other analytes. In this commentary, we review the current evidence on these tests, provide several major considerations for new MCED tests, and outline how their evaluation will need to differ from that established for traditional single-cancer screening tests.
Colorectal cancer is the third most common cancer worldwide. Because of the slow progression of the precancerous precursors, an efficient endoscopic surveillance strategy may be expected. It seems that around one-fourth of colorectal malignancies are still missed during colonoscopy. Several endoscopic technologies have been introduced, without radical changes. Interest in the development of artificial intelligence applications in the medical field has grown in the past decade. Artificial intelligence can help to highlight a specific region of interest that needs closer examination for the identification of polyps. The aim of this review is to report the first clinical experiences with the first US FDA-approved, real-time, deep-learning, computer-aided detection system (GI Genius™, Medtronic).
Artificial intelligence (AI) applications for screening mammography are being marketed for clinical use in the interpretative domains of lesion detection and diagnosis, triage, and breast density assessment, and in the noninterpretive domains of breast cancer risk assessment, image quality control, image acquisition, and dose reduction. Evidence in support of these nascent applications, particularly for lesion detection and diagnosis, is largely based on multireader studies with cancer-enriched datasets rather than rigorous clinical evaluation aligned with the application's specific intended clinical use. This article reviews commercial AI algorithms for screening mammography that are currently available for clinical practice, their use, and evidence supporting their performance. Clinical implementation considerations, such as workflow integration, governance, and ethical issues, are also described. In addition, the future of AI for screening mammography is discussed, including the development of interpretive and noninterpretive AI applications and strategic priorities for research and development.
Advances in quantitative biomarker development have accelerated new forms of data-driven insights for patients with cancer. However, most approaches are limited to a single mode of data, leaving integrated approaches across modalities relatively underdeveloped. Multimodal integration of advanced molecular diagnostics, radiological and histological imaging, and codified clinical data presents opportunities to advance precision oncology beyond genomics and standard molecular techniques. However, most medical datasets are still too sparse to be useful for the training of modern machine learning techniques, and significant challenges remain before this is remedied. Combined efforts of data engineering, computational methods for analysis of heterogeneous data and instantiation of synergistic data models in biomedical research are required for success. In this Perspective, we offer our opinions on synthesizing complementary modalities of data with emerging multimodal artificial intelligence methods. Advancing along this direction will result in a reimagined class of multimodal biomarkers to propel the field of precision oncology in the coming decade. This Perspective proposes that data from multiple modalities, including molecular diagnostics, radiological and histological imaging and codified clinical data, should be integrated by multimodal machine learning models to advance the prognosis and treatment management of patients with cancer.
AI algorithms used for diagnosis and prognosis must be explainable and must not rely on a black box.
This paper offers a summary of the latest studies on healthcare scheduling problems including patients' admission scheduling problem, nurse scheduling problem, operation room scheduling problem, surgery scheduling problem and other healthcare scheduling problems. The paper provides a comprehensive survey on healthcare scheduling focuses on the recent literature. The development of healthcare scheduling research plays a critical role in optimizing costs and improving the patient flow, providing prompt administration of treatment, and the optimal use of the resources provided and accessible in the hospitals. In the last decades, the healthcare scheduling methods that aim to automate the search for optimal resource management in hospitals by using metaheuristics methods have proliferated. However, the reported results are disintegrated since they solved every specific problem independently, given that there are many versions of problem definition and various data sets available for each of these problems. Therefore, this paper integrates the existing results by performing a comprehensive review and analyzing 190 articles based on four essential components in solving optimization problems: problem definition, formulations, data sets, and methods. This paper summarizes the latest healthcare scheduling problems focusing on patients' admission scheduling problems, nurse scheduling problems, and operation room scheduling problems considering these are the most common issues found in the literature. Furthermore, this review aims to help researchers to highlight some development from the most recent papers and grasp the new trends for future directions.
Artificial intelligence (AI) is rapidly reshaping cancer research and personalized clinical care. Availability of high-dimensionality datasets coupled with advances in high-performance computing, as well as innovative deep learning architectures, has led to an explosion of AI use in various aspects of oncology research. These applications range from detection and classification of cancer, to molecular characterization of tumors and their microenvironment, to drug discovery and repurposing, to predicting treatment outcomes for patients. As these advances start penetrating the clinic, we foresee a shifting paradigm in cancer care becoming strongly driven by AI.
Significance: AI has the potential to dramatically affect nearly all aspects of oncology—from enhancing diagnosis to personalizing treatment and discovering novel anticancer drugs. Here, we review the recent enormous progress in the application of AI to oncology, highlight limitations and pitfalls, and chart a path for adoption of AI in the cancer clinic.
Objectives
Assessment of patient-reported outcomes (PROs) in oncology is of critical importance because it provides unique information that may also predict clinical outcomes.
Methods
We conducted a systematic review of prognostic factor studies to examine the prognostic value of PROs for survival in cancer. A systematic literature search was performed in PubMed for studies published between 2013 and 2018. We considered any study, regardless of the research design, that included at least 1 PRO domain in the final multivariable prognostic model. The protocol (EPIPHANY) was published and registered in the International Prospective Register of Systematic Reviews (CRD42018099160).
Results
Eligibility criteria selected 138 studies including 158 127 patients, of which 43 studies were randomized, controlled trials. Overall, 120 (87%) studies reported at least 1 PRO to be statistically significantly prognostic for overall survival. Lung (n = 41, 29.7%) and genitourinary (n = 27, 19.6%) cancers were most commonly investigated. The prognostic value of PROs was investigated in secondary data analyses in 101 (73.2%) studies. The EORTC QLQ-C30 questionnaire was the most frequently used measure, and its physical functioning scale (range 0-100) the most frequent independent prognostic PRO, with a pooled hazard ratio estimate of 0.88 per 10-point increase (95% CI 0.84-0.92).
Conclusions
There is convincing evidence that PROs provide independent prognostic information for overall survival across cancer populations and disease stages. Further research is needed to translate current evidence-based data into prognostic tools to aid in clinical decision making.
This review discusses the recent advances in the accelerated search for biologic meaning of radiomics signatures, as the biologic validation is gradually recognized as essential for the field to enter routine clinical practice.
Radiomic analysis offers a powerful tool for the extraction of clinically relevant information from radiologic imaging. Radiomics can be used to predict patient outcome through automated high-throughput feature extraction, using large training cohorts to elucidate subtle relationships between image characteristics and disease status. However powerful, the data-driven nature of radiomics inherently offers no insight into the biological underpinnings of the observed relationships. Early radiomics work was dominated by analysis of semantic, radiologist-defined features and carried qualitative real-world meaning. Following the rapid developments and popularity of machine learning approaches, the field moved quickly toward high-throughput agnostic analyses, resulting in increasingly large feature sets. This trend took the focus toward an increase in predictive power and further away from a biological understanding of the findings. Such a disconnect between predictor model and biological meaning will inherently limit broad clinical translation. Efforts to reintroduce biological meaning into radiomics are gaining traction in the field with distinct emerging approaches available, including genomic correlates, local microscopic pathologic image textures, and macroscopic histopathologic marker expression. These methods are presented in this review, and their significance is discussed. The authors predict that following the increasing pressure for robust radiomics, biological validation will become a standard practice in the field, thus further cementing the role of the method in clinical decision making.
PURPOSE
Because of expanding interoperability requirements, structured patient data are increasingly available in electronic health records. Many oncology data elements (eg, staging, biomarkers, documentation of adverse events and cancer outcomes) remain challenging. The Minimal Common Oncology Data Elements (mCODE) project is a consensus data standard created to facilitate transmission of data of patients with cancer.
METHODS
In 2018, mCODE was developed through a work group convened by ASCO, including oncologists, informaticians, researchers, and experts in terminologies and standards. The mCODE specification is organized by 6 high-level domains: patient, laboratory/vital, disease, genomics, treatment, and outcome. In total, 23 mCODE profiles are composed of 90 data elements.
RESULTS
A conceptual model was published for public comment in January 2019 and, after additional refinement, the first public version of the mCODE (version 0.9.1) Fast Healthcare Interoperability Resources (FHIR) implementation guide (IG) was presented at the ASCO Annual Meeting in June 2019. The specification was approved for balloting by Health Level 7 International (HL7) in August 2019. mCODE passed the HL7 ballot in September 2019 with 86.5% approval. The mCODE IG authors worked with HL7 reviewers to resolve all negative comments, leading to a modest expansion in the number of data elements and tighter alignment with FHIR and other HL7 conventions. The mCODE version 1.0 FHIR IG Standard for Trial Use was formally published on March 18, 2020.
CONCLUSION
The mCODE project has the potential to offer tremendous benefits to cancer care delivery and research by creating an infrastructure to better share patient data. mCODE is available free from www.mCODEinitiative.org . Pilot implementations are underway, and a robust community of stakeholders has been assembled across the oncology ecosystem.
PURPOSE
Cancer research using electronic health records and genomic data sets requires clinical outcomes data, which may be recorded only in unstructured text by treating oncologists. Natural language processing (NLP) could substantially accelerate extraction of this information.
METHODS
Patients with lung cancer who had tumor sequencing as part of a single-institution precision oncology study from 2013 to 2018 were identified. Medical oncologists’ progress notes for these patients were reviewed. For each note, curators recorded whether the assessment/plan indicated any cancer, progression/worsening of disease, and/or response to therapy or improving disease. Next, a recurrent neural network was trained using unlabeled notes to extract the assessment/plan from each note. Finally, convolutional neural networks were trained on labeled assessments/plans to predict the probability that each curated outcome was present. Model performance was evaluated using the area under the receiver operating characteristic curve (AUROC) among a held-out test set of 10% of patients. Associations between curated response or progression end points and overall survival were measured using Cox models among patients receiving palliative-intent systemic therapy.
RESULTS
Medical oncologist notes (n = 7,597) were manually curated for 919 patients. In the 10% test set, NLP models replicated human curation with AUROCs of 0.94 for the any-cancer outcome, 0.86 for the progression outcome, and 0.90 for the response outcome. Progression/worsening events identified using NLP models were associated with shortened survival (hazard ratio [HR] for mortality, 2.49; 95% CI, 2.00 to 3.09); response/improvement events were associated with improved survival (HR, 0.45; 95% CI, 0.30 to 0.67).
CONCLUSION
NLP models based on neural networks can extract meaningful outcomes from oncologist notes at scale. Such models may facilitate identification of clinical and genomic features associated with response to cancer treatment.
Artificial intelligence (AI) was first described in 1950; however, several limitations in early models prevented widespread acceptance and application to medicine. In the early 2000s, many of these limitations were overcome by the advent of deep learning. Now AI systems are capable of analyzing complex algorithms and self-learning, we enter a new age in medicine where AI can be applied to clinical practice through risk assessment models, improving diagnostic accuracy and improving workflow efficiency. This article presents a brief historical perspective on the evolution of AI over the last several decades and the introduction and development of AI in medicine in recent years. A brief summary of the major applications of AI in gastroenterology and endoscopy are also presented, which will be reviewed in further detail by several other articles in this issue of GIE.
As patient data are increasingly captured digitally, the opportunities to deploy artificial intelligence (AI), especially machine learning, are increasing rapidly. Machine learning is automated learning by computers using tools such as artificial neural networks to search data iteratively for optimal solutions.¹ Typical applications include searching for novel patterns (eg, latent cancer subtypes²), making a diagnosis or outcome prediction (eg, diabetic retinopathy³), and optimizing treatment decisions (eg, fluid and vasopressor titration for septic shock⁴). Although many express excitement regarding the promise of AI, others express concern about adverse consequences, such as loss of physician and patient autonomy or unintended bias, and still others claim that the entire endeavor is largely hype, with virtually no data that actual patient outcomes have improved.⁵,6
The promise of artificial intelligence (AI) in health care offers substantial opportunities to improve patient and clinical team outcomes, reduce costs, and influence population health. Current data generation greatly exceeds human cognitive capacity to effectively manage information, and AI is likely to have an important and complementary role to human cognition to support delivery of personalized health care.¹ For example, recent innovations in AI have shown high levels of accuracy in imaging and signal detection tasks and are considered among the most mature tools in this domain.
Recent scrutiny of artificial intelligence (AI)–based facial recognition software has renewed concerns about the unintended effects of AI on social bias and inequity. Academic and government officials have raised concerns over racial and gender bias in several AI-based technologies, including internet search engines and algorithms to predict risk of criminal behavior. Companies like IBM and Microsoft have made public commitments to “de-bias” their technologies, whereas Amazon mounted a public campaign criticizing such research. As AI applications gain traction in medicine, clinicians and health system leaders have raised similar concerns over automating and propagating existing biases.
Racial bias in health algorithms
The U.S. health care system uses commercial algorithms to guide health decisions. Obermeyer et al. find evidence of racial bias in one widely used algorithm, such that Black patients assigned the same level of risk by the algorithm are sicker than White patients (see the Perspective by Benjamin). The authors estimated that this racial bias reduces the number of Black patients identified for extra care by more than half. Bias occurs because the algorithm uses health costs as a proxy for health needs. Less money is spent on Black patients who have the same level of need, and the algorithm thus falsely concludes that Black patients are healthier than equally sick White patients. Reformulating the algorithm so that it no longer uses costs as a proxy for needs eliminates the racial bias in predicting who needs extra care.
Science , this issue p. 447 ; see also p. 421
There is currently no other hot topic like the ability of current technology to develop capabilities similar to those of human beings, even in medicine. This ability to simulate the processes of human intelligence with computer systems is known as artificial intelligence (AI). This article aims to clarify the various terms that still sound foreign to us, such as AI, machine learning (ML), deep learning (DL), and big data. It also provides an in-depth description of the concept of AI and its types; the learning techniques and technology used by ML; cardiac imaging analysis with DL; and the contribution of this technological revolution to classical statistics, as well as its current limitations, legal aspects, and initial applications in cardiology. To do this, we conducted a detailed PubMed search on the evolution of original contributions on AI to the various areas of application in cardiology in the last 5 years and identified 673 research articles. We provide 19 detailed examples from distinct areas of cardiology that, by using AI, have shown diagnostic and therapeutic improvements, and which will aid understanding of ML and DL methodology.
In the past decade, advances in precision oncology have resulted in an increased demand for predictive assays that enable the selection and stratification of patients for treatment. The enormous divergence of signalling and transcriptional networks mediating the crosstalk between cancer, stromal and immune cells complicates the development of functionally relevant biomarkers based on a single gene or protein. However, the result of these complex processes can be uniquely captured in the morphometric features of stained tissue specimens. The possibility of digitizing whole-slide images of tissue has led to the advent of artificial intelligence (AI) and machine learning tools in digital pathology, which enable mining of subvisual morphometric phenotypes and might, ultimately, improve patient management. In this Perspective, we critically evaluate various AI-based computational approaches for digital pathology, focusing on deep neural networks and ‘hand-crafted’ feature-based methodologies. We aim to provide a broad framework for incorporating AI and machine learning tools into clinical oncology, with an emphasis on biomarker development. We discuss some of the challenges relating to the use of AI, including the need for well-curated validation datasets, regulatory approval and fair reimbursement strategies. Finally, we present potential future opportunities for precision oncology. The authors of this Perspective critically evaluate various artificial intelligence (AI)-based computational approaches used for digital pathology and provide a broad framework to incorporate these tools into clinical oncology, discussing challenges such as the need for well-curated validation datasets, regulatory approval and fair reimbursement strategies.
Background:
Atrial fibrillation is frequently asymptomatic and thus underdetected but is associated with stroke, heart failure, and death. Existing screening methods require prolonged monitoring and are limited by cost and low yield. We aimed to develop a rapid, inexpensive, point-of-care means of identifying patients with atrial fibrillation using machine learning.
Methods:
We developed an artificial intelligence (AI)-enabled electrocardiograph (ECG) using a convolutional neural network to detect the electrocardiographic signature of atrial fibrillation present during normal sinus rhythm using standard 10-second, 12-lead ECGs. We included all patients aged 18 years or older with at least one digital, normal sinus rhythm, standard 10-second, 12-lead ECG acquired in the supine position at the Mayo Clinic ECG laboratory between Dec 31, 1993, and July 21, 2017, with rhythm labels validated by trained personnel under cardiologist supervision. We classified patients with at least one ECG with a rhythm of atrial fibrillation or atrial flutter as positive for atrial fibrillation. We allocated ECGs to the training, internal validation, and testing datasets in a 7:1:2 ratio. We calculated the area under the curve (AUC) of the receiver operatoring characteristic curve for the internal validation dataset to select a probability threshold, which we applied to the testing dataset. We evaluated model performance on the testing dataset by calculating the AUC and the accuracy, sensitivity, specificity, and F1 score with two-sided 95% CIs.
Findings:
We included 180 922 patients with 649 931 normal sinus rhythm ECGs for analysis: 454 789 ECGs recorded from 126 526 patients in the training dataset, 64 340 ECGs from 18 116 patients in the internal validation dataset, and 130 802 ECGs from 36 280 patients in the testing dataset. 3051 (8·4%) patients in the testing dataset had verified atrial fibrillation before the normal sinus rhythm ECG tested by the model. A single AI-enabled ECG identified atrial fibrillation with an AUC of 0·87 (95% CI 0·86-0·88), sensitivity of 79·0% (77·5-80·4), specificity of 79·5% (79·0-79·9), F1 score of 39·2% (38·1-40·3), and overall accuracy of 79·4% (79·0-79·9). Including all ECGs acquired during the first month of each patient's window of interest (ie, the study start date or 31 days before the first recorded atrial fibrillation ECG) increased the AUC to 0·90 (0·90-0·91), sensitivity to 82·3% (80·9-83·6), specificity to 83·4% (83·0-83·8), F1 score to 45·4% (44·2-46·5), and overall accuracy to 83·3% (83·0-83·7).
Interpretation:
An AI-enabled ECG acquired during normal sinus rhythm permits identification at point of care of individuals with atrial fibrillation.
Funding:
None.
Resumen
Existen pocos temas de actualidad equiparables a la posibilidad de la tecnología actual para desarrollar las mismas capacidades que el ser humano, incluso en medicina. Esta capacidad de simular los procesos de inteligencia humana por parte de máquinas o sistemas informáticos es lo que conocemos hoy en día como inteligencia artificial (IA). Este artículo pretende aclarar diferentes términos que todavía nos resultan lejanos como IA, machine learning (aprendizaje automático, AA), deep learning (aprendizaje profundo, AP), data science o big data; describir en profundidad el concepto de IA y sus tipos, las técnicas de aprendizaje y la metodología que se utiliza en el AA, el análisis en imagen cardiaca con AP, la aportación de esta revolución tecnológica a la estadística clásica, sus limitaciones actuales, sus aspectos legales y, fundamentalmente, sus aplicaciones iniciales en cardiología. En este sentido se ha realizado una búsqueda detallada en PubMed de la evolución en el último lustro de las contribuciones de la IA a las diferentes áreas de aplicación en cardiología, y se ha identificado un total de 673 artículos originales. Se describen en detalle 19 ejemplos de diferentes áreas de la cardiología que utilizando IA han mostrado mejoras diagnósticas y terapéuticas, y que facilitarán la comprensión de la metodología AA y AP.
Bringing together artificial intelligence and neuroscience promises to yield benefits for both fields. Bringing together artificial intelligence and neuroscience promises to yield benefits for both fields.