No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
A Novel Transfer Learning Based Approach for Pneumonia Detection in Chest X-ray Images
Abstract
Pneumonia is among the top diseases which cause most of the deaths all over the world. Virus, bacteria and fungi can all cause pneumonia. However, it is difficult to judge the pneumonia just by looking at chest X-rays. The aim of this study is to simplify the pneumonia detection process for experts as well as for novices. We suggest a novel deep learning framework for the detection of pneumonia using the concept of transfer learning. In this approach, features from images are extracted using different neural network models pretrained on ImageNet, which then are fed into a classifier for prediction. We prepared five different models and analyzed their performance. Thereafter, we proposed an ensemble model that combines outputs from all pretrained models, which outperformed individual models, reaching the state-of-the-art performance in pneumonia recognition. Our ensemble model reached an accuracy of 96.4% with a recall of 99.62% on unseen data from the Guangzhou Women and Children's Medical Center dataset.
... Transfer learning for feature extraction and machine learning for classification Due to the small size of the dataset in this experiment, combined with a particular depth of the deep neural network, makes it susceptible to overfitting during the training process. This, in turn, results in a poor ability to accurately recognize patterns 34,35 . In this scenario, the pretrained convolutional neural network may be employed to extract previously acquired image features. ...
... So, in this study, we employed VGG19, Resnet18, and MobileNetV2 as feature extractor which are the most basic, and make modifications to adapt it to our cancer classification task and DT, k-NN, Naïve Bayes and SVM as classifiers. All of the aforementioned pre-trained networks utilized the weight parameters from the ImageNet dataset 35 which contains more than a million images of different categories. The following sub section provides the details of each pre-trained network models. ...
Accurately and early diagnosis of melanoma is one of the challenging tasks due to its unique characteristics and different shapes of skin lesions. So, in order to solve this issue, the current study examines various deep learning-based approaches and provide an effective approach for classifying dermoscopic images into two categories of skin lesions. This research focus on skin cancer images and provides solution using deep learning approaches. This research investigates three approaches for classifying skin cancer images. (1) Utilizing three fine-tuned pre-trained networks (VGG19, ResNet18, and MobileNet_V2) as classifiers. (2) Employing three pre-trained networks (ResNet-18, VGG19, and MobileNet v2) as feature extractors in conjunction with four machine learning classifiers (SVM, DT, Naïve Bayes, and KNN). (3) Utilizing a combination of the aforementioned pre-trained networks as feature extractors in conjunction with same machine learning classifiers. All these algorithms are trained using segmented images which are achieved by using the active contour approach. Prior to segmentation, preprocessing step is performed which involves scaling, denoising, and enhancing the image. Experimental performance is measured on the ISIC 2018 dataset which contains 3300 images of skin disease including benign and malignant type cancer images. 80% of the images from the ISIC 2018 dataset are allocated for training, while the remaining 20% are designated for testing. All approaches are trained using different parameters like epoch, batch size, and learning rate. The results indicate that combining ResNet-18 and MobileNet pre-trained networks using concatenation with an SVM classifier achieved the maximum accuracy of 92.87%.
... Feature selection was performed using the mRMR method, and the best results were achieved by combining all features from mRMR, with an accuracy of 99.41% using the Kermany dataset (70% training, 30% testing) [27]. Chouhan et al. proposed an ensemble model that combined the outputs from several pre-trained models, surpassing the performance of individual models [28]. ...
... This ensemble, which included Inception V3, ResNet, AlexNet, GoogleNet, and DenseNet121, achieved state-ofthe-art results with an accuracy of 96.4% and a recall of 99.62% on the Kermany dataset, using a 90% training and 10% testing split [28]. Hashmi et al. introduced a novel weighted classifier approach that optimally combined predictions from several state-of-the-art deep learning models, including ResNet18, Xception, InceptionV3, DenseNet121, and MobileNetV3 [29]. ...
Pneumonia is a leading cause of illness and death in children, underscoring the need for early and accurate detection. In this study, we propose a novel lightweight ensemble model for detecting pneumonia in children using chest X-ray images. This ensemble model integrates two pre-trained convolutional neural networks (CNNs), MobileNetV2 and NASNetMobile, selected for their balance of computational efficiency and accuracy. These models were fine-tuned on a pediatric chest X-ray dataset and combined to enhance classification performance. Our proposed ensemble model achieved a classification accuracy of 98.63%, significantly outperforming individual models such as MobileNetV2 (97.10%) and NASNetMobile(96.25%) in terms of accuracy, precision, recall, and F1 score. Moreover, the ensemble model outperformed state-of-the-art architectures, including ResNet50, InceptionV3, and DenseNet201, while maintaining computational efficiency. The proposed lightweight ensemble model presents a highly effective and resource-efficient solution for pneumonia detection, making it particularly suitable for deployment in resource-constrained settings.
... The emergence of machine learning, particularly Convolutional Neural Networks (CNNs), has offered promising avenues for automating the detection of pneumonia from chest X-ray images [6]. Conventional methods for pneumonia diagnosis heavily rely on manual interpretation of medical imaging by trained radiologists [7,8]. However, this process is time-consuming, subjective, and may suffer from interobserver variability. ...
In recent years, the integration of machine learning techniques within the medical field has shown promising results in aiding healthcare professionals in accurate diagnosis and treatment planning. This study focuses on developing and implementing a machine learning model tailored specifically for medical diagnosis, leveraging advancements in computer vision and deep learning algorithms. This research aims to design an efficient and accurate model capable of classifying medical images into distinct categories, enabling automated diagnosis and identification of various ailments and conditions. This study uses a dataset comprising 5,863 Chest X-ray images (JPEG) and 2 categories (Pneumonia/Normal) (anterior-posterior) selected from retrospective cohorts of pediatric patients of one to five years old from Guangzhou Women and Children's Medical Center, Guangzhou, obtained from Kaggle data repositories. Data Preprocessing was conducted to enhance image quality and extract relevant features, followed by implementing a deep convolutional neural networks (DCNNs) model using TensorFlow's Keras. Using pre-trained models such as Resnet, transfer learning techniques were employed to learn efficient features from large-scale datasets and optimize the model's performance with the limited medical data available. The results from the experimental analysis showed that after 9 epochs, the training and validation accuracies had steadily increased, achieving 95% and 75%, respectively. Overall, the model achieved 99.9% training accuracy across multiple epochs and an average validation accuracy of 75%. The model's performance and scalability highlight its potential for integration into clinical workflows. This could revolutionize healthcare by augmenting the diagnostic process and improving patient outcomes.
... Therefore, we collected around 30,000 medical images from the open literature before applying the proposed noise-synthesizing algorithm. The collected data samples include multi-modal medical images acquired with different acquisition devices and datasets: i) MRI [28], [29], ii) X-ray [30], [31]), iii) CT images [32], [33], iv) Skin lesion images [34]), and v) Microscopy (i.e., protein atlas [35], [36]). We used 24,000 random images for training and 1,000 images for validation. ...
Medical image denoising is considered among the most challenging vision tasks. Despite the real-world implications, existing denoising methods have notable drawbacks as they often generate visual artifacts when applied to heterogeneous medical images. This study addresses the limitation of the contemporary denoising methods with an artificial intelligence (AI)-driven two-stage learning strategy. The proposed method learns to estimate the residual noise from the noisy images. Later, it incorporates a novel noise attention mechanism to correlate estimated residual noise with noisy inputs to perform denoising in a course-to-refine manner. This study also proposes to leverage a multi-modal learning strategy to generalize the denoising among medical image modalities and multiple noise patterns for widespread applications. The practicability of the proposed method has been evaluated with dense experiments. The experimental results demonstrated that the proposed method achieved state-of-the-art performance by significantly outperforming the existing medical image denoising methods in quantitative and qualitative comparisons. Overall, it illustrates a performance gain of 7.64 in Peak Signal-to-Noise Ratio (PSNR), 0.1021 in Structural Similarity Index (SSIM), 0.80 in DeltaE (), 0.1855 in Visual Information Fidelity Pixel-wise (VIFP), and 18.54 in Mean Squared Error (MSE) metrics.
... Previous study has demonstrated potential in developing AI-based methods for GGO segmentation [43]. Different algorithms, such as those in radiomics and deep learning, are making significant contributions; these algorithms hold promise not only in differentiating benign from malignant nodules but also in predicting the prognosis of small-cell lung cancer and pneumonia cases [44][45][46][47][48]. However, contradictory results were obtained regarding the performance of AI in GGO screening and diagnosis, with some studies reporting poor performance and others reporting better performance compared to traditional methods. ...
Ground-glass opacities (GGOs) are hazy opacities on chest computed tomography (CT) scans that can indicate various lung diseases, including early COVID-19, pneumonia, and lung cancer. Artificial intelligence (AI) is a promising tool for analyzing medical images, such as chest CT scans. The aim of this study was to evaluate AI models' performance in detecting GGO nodules using metrics like accuracy, sensitivity, specificity, F1 score, area under the curve (AUC) and precision. We designed a search strategy to include reports focusing on deep learning algorithms applied to high-resolution CT scans. The search was performed on PubMed, Google Scholar, Scopus, and ScienceDirect to identify studies published between 2016 and 2024. Quality appraisal of included studies was conducted using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool, assessing the risk of bias and applicability concerns across four domains. Two reviewers independently screened studies reporting the diagnostic ability of AI-assisted CT scans in early GGO detection, where the review results were synthesized qualitatively. Out of 5,247 initially identified records, we found 18 studies matching the inclusion criteria of this study. Among evaluated models, DenseNet achieved the highest accuracy of 99.48%, though its sensitivity and specificity were not reported. WOANet showed an accuracy of 98.78%, with a sensitivity of 98.37% and high specificity of 99.19%, excelling particularly in specificity without compromising sensitivity. In conclusion, AI models can potentially detect GGO on chest CT scans. Future research should focus on developing hybrid models that integrate various AI approaches to improve clinical applicability.
... Their approach involved image segmentation, model training, feature extraction, and classification using Gaussian filters, yielding an AUC of 0.85 and corresponding sensitivity and specificity rates of 76% and 80%. In a related study, Jain et al. [27] trained six CNN models on pediatric CXR to distinguish between pneumonia and non-pneumonia cases, including VGG16, VGG19, Inception v3, ResNet50, and two and three-layer CNN, optimized with various parameters, attained an accuracy of 92.31%. Chouhan et al. [28] achieved a remarkable 96.4% accuracy in pneumonia diagnosis by employing a hybrid ensemble of top-performing models to analyze CXR data from the GWCMC. ...
bold xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Objective: Neonatal Respiratory Distress Syndrome (NRDS) poses a significant threat to newborn health, necessitating timely and accurate diagnosis. This study introduces NDL-Net, an innovative hybrid deep learning framework designed to diagnose NRDS from chest X-rays (CXR). Results: The architecture combines MobileNetV3 Large for efficient image processing and ResNet50 for detecting complex patterns essential for NRDS identification. Additionally, a Long Short-Term Memory (LSTM) layer analyzes temporal variations in imaging data, enhancing predictive accuracy. Extensive evaluation on neonatal CXR datasets demonstrated NDL-Net's high diagnostic performance, achieving 98.09% accuracy, 97.45% precision, 98.73% sensitivity, 98.08% F1-score, and 98.73% specificity. The model's low false negative and false positive rates underscore its superior diagnostic capabilities. Conclusion: NDL-Net represents a significant advancement in medical diagnostics, improving neonatal care through early detection and management of NRDS.
... The development of automated systems such as the use of artificial intelligence (AI) and machine learning (ML) algorithms can precisely detect and diagnose pneumonia based on the available medical images. The history of automated diagnosis of pneumonia may be traced back to the development of digital medical imaging technologies such as X-rays, magnetic resonance imaging (MRI), and computed tomography (CT) [8,3]. Clinicians were able to see the lungs and spot problems like pneumonia because of imaging technologies. ...
A chest x-ray is considered to be a readily available and a reliable and fast method for the diagnosis of the pneumonia. In literature, different object detection algorithm are used to the x-ray images. Since, pneumonia can be highly life threatening, the decisions made by the object detection model can influence the lives of thousands of people. It is imperative that we precisely know the performance and the accuracy of the model so that the stakeholders have a wider view of how trustworthy the results are. There is a need to understand the performance and accuracy of the underlying deep learning model. The architecture of the model such as number of neural layers or the computation cost poses a great impact on the accuracy and efficiency. In this paper, our contribution is twofold. Firstly, we have proposed a new deep learning (DL) strategy for the automated identification of pneumonia employing a convolutional neural network (CNN) with improved accuracy. We have introduced a hybrid model in which we have used a pre-trained CNN model, a vision transformer (ViT) and applied N times and N number of CNN layers to improve the accuracy and computational performance. We have identified a point where the accuracy of the model is highest and computational cost is acceptable. Secondly, we have compared our proposed model with different other classifiers such as You Only Look Once (YOLO) v4, and CNN and its multiple layers. Using a publicly accessible dataset, this study attempted to pre-process the input chest X-ray images in order to detect the presence of pneumonia. The experimental results have revealed that the accuracy level of the proposed hybrid CNN model reached a level of 97.6% on datasets of varying sizes.
... Chowdhury, M. E. H. et al. investigated the role of AI in screening viral and COVID-19 pneumonia, highlighting the broader applications of AI in pneumonia diagnosis (Chowdhury et al., 2020). Chouhan, V. et al. introduced a novel transfer learning-based approach for pneumonia detection in chest X-ray images, demonstrating the continuous evolution of transfer learning in pneumonia diagnosis (Chouhan et al., 2020). Mahmud, T. et al. presented CovXNet, a multi-dilation convolutional neural network for automatic COVID-19 and pneumonia detection from chest X-ray images, showcasing innovative approaches to pneumonia diagnosis (Mahmud et al., 2020). ...
This research, dedicated to developing an accurate and efficient pneumonia detection system from Chest X-Ray images, highlights the significance of automated tools in enhancing healthcare diagnostics. Its significance lies in the fact that pneumonia is a prevalent respiratory condition that requires timely and accurate diagnosis for effective medical intervention. The project's objective was to make use of convolutional neural networks and image analyses to create an automated diagnostic tool that could assist healthcare professionals in identifying pneumonia with precision and efficiency. To achieve this, the system initially made use of two custom deep learning architectures but ultimately used a pretrained CheXNet-based model, developed by using transfer learning. This choice was made by considering CheXNet’s proven performance in identifying pneumonia and other pulmonary conditions. The project's results proved promising, with the CheXNet-based model achieving high diagnostic accuracy and providing valuable insights into the presence of pneumonia. The system's architecture, using deep learning and the use of DICOM images, demonstrated its effectiveness in improving the accuracy and efficiency of pneumonia diagnosis. Based on the results, this paper further demonstrates a web-based application for interaction with the system. Additionally, it provides information on the work that could be done in the future. Thus, this research contributes to the growing field of medical image analysis and highlights the significance of automated tools in enhancing healthcare diagnostics. The project's outcomes are meant to pave the way for more efficient and accessible methods for pneumonia detection, ultimately benefiting both healthcare providers and patients. KEYWORDS: Pneumonia, Chest X-rays, Diagnostic Support System, Machine Learning, CheXNet, DICOM.
... Machine learning can solve complex computer vision problems in medical imaging [10]. However, deep learning has taken giant steps in these models because of its ability to learn hierarchical features from raw image data [11]. ...
Pneumonia is a deadly disease affecting millions worldwide, caused by microorganisms and environmental factors. It leads to lung fluid build-up, making breathing difficult, and is a leading cause of death. Early detection and treatment are crucial for preventing severe outcomes. Chest X-rays are commonly used for diagnoses due to their accessibility and low costs; however, detecting pneumonia through X-rays is challenging. Automated methods are needed, and machine learning can solve complex computer vision problems in medical imaging. This research develops a robust machine learning model for the early detection of pneumonia using chest X-rays, leveraging advanced image processing techniques and deep learning algorithms that accurately identify pneumonia patterns, enabling prompt diagnosis and treatment. The research develops a CNN model from the ground up and a ResNet-50 pretrained model This study uses the RSNA pneumonia detection challenge original dataset comprising 26,684 chest array images collected from unique patients (56% male, 44% females) to build a machine learning model for the early detection of pneumonia. The data are made up of pneumonia (31.6%) and non-pneumonia (68.8%), providing an effective foundation for the model training and evaluation. A reduced size of the dataset was used to examine the impact of data size and both versions were tested with and without the use of augmentation. The models were compared with existing works, the model’s effectiveness in detecting pneumonia was compared with one another, and the impact of augmentation and the dataset size on the performance of the models was examined. The overall best accuracy achieved was that of the CNN model from scratch, with no augmentation, an accuracy of 0.79, a precision of 0.76, a recall of 0.73, and an F1 score of 0.74. However, the pretrained model, with lower overall accuracy, was found to be more generalizable.
... A novel deep learning method for the identification of pneumonia by mobile learning was proposed utilizing X-ray images from the Maternal and Child Medical Center dataset [22]. AlexNet (MAN) was used to improve the categorization of chest radiographs [23]. ...
COVID-19 pneumonia is a persistent worldwide health problem that usually affects the most vulnerable groups in society: the newborn and aged populations. Most of the current endeavors toward handling diagnosis and classification of pneumonia have used numerous techniques for machine learning and deep learning, with a particular focus on COVID-19 pneumonia. However, most of these techniques have raised concerns with regard to diagnostic precision as a result of the limited application of advanced algorithms, datasets whose validation is mostly inadequate and predictive capability. To address these limitations, our research introduces a deep learning-based approach by Convolutional Neural Networks (CNNs), which enhances the performance in classifying COVID-19 pneumonia. Salient features of the proposed method include a four-step process: first, data acquisition from a comprehensive chest X-ray dataset on GitHub; second, processing and analyzing the data through visual means like histograms and scatter plots; third, using CNNs supplemented with techniques for data augmentation in supervised learning; lastly, performance evaluation to benchmark against existing models. The present study uses features from X-ray images with the help of CNN's advanced pattern recognition capabilities in pursuit of achieving better generalization and precision in classification. The model achieved an accuracy of 85.70\% and precision of 88.6%, which surpasses the existing techniques and thereby built the promise of improving the accuracy of the diagnostic process, hence, leading to improved care for the patients, and more so forms the foundation on which future AI-powered healthcare diagnostics are expected to take off.
... The dataset we chose was very limited in images (containing only a thousand of each type), so as to check the performance of the model with constraints. Furthermore, some of the images lack quality and are a bit hazy which would provide a more challenging environment for the model [21][22][23][24][25][26]. ...
... We have also compared the performance of various models and the state-of-the-art outputs in other similar reported studies, as detailed in Table 6. Using the ensemble model (ResNet18, AlexNet, Inception v3, GoogleNet, and Den-seNet121), Chouhan et al. [37] Furthermore, Apostolopoulos and Mpesiana [43] obtained 96.78% accuracy, 98.66% specificity, and 96.46 sensitivity using transfer learning approaches to analyse pneumonia, COVID-19, and normal lung CXRs. ConvMixer, the proposed model under the present study, returned the best accuracies (97% for multiclass and 98% for binary class, with combined augmentation techniques) as compared to other similar stand-alone state-of-the-art models. ...
Background
The highly infectious coronavirus disease 2019 (COVID-19) is caused by severe acute respiratory syndrome coronavirus 2, the seventh coronavirus. It is the longest pandemic in recorded history worldwide. Many countries are still reporting COVID-19 cases even in the fifth year of its emergence.
Objective
The performance of various machine learning (ML) and deep learning (DL) models was studied for image-based classification of the lungs infected with COVID-19, pneumonia (viral and bacterial), and normal cases from the chest X-rays (CXRs).
Methods
The K-nearest neighbour and logistics regression as the two ML models, and Visual Geometry Group-19, Vision transformer, and ConvMixer as the three DL models were included in the investigation to compare the brevity of the detection and classification of the cases.
Results
Among the investigated models, ConvMixer returned the best result in terms of accuracy, recall, precision, F1-score and area under the curve for both binary as well as multiclass classification. The pre-trained ConvMixer model outperformed the other four models in classifying. As per the performance observations, there was 97.1% accuracy for normal and COVID-19 + pneumonia-infected lungs, 98% accuracy for normal and COVID-19 infected lungs, 82% accuracy for normal + bacterial + viral infected lungs, and 98% accuracy for normal + pneumonia infected lungs. The DL models performed better than the ML models for binary and multiclass classification. The performance of these studied models was tried on other CXR image databases.
Conclusion
The suggested network effectively detected COVID-19 and different types of pneumonia by using CXR imagery. This could help medical sciences for timely and accurate diagnoses of the cases through bioimaging technology and the use of high-end bioinformatics tools.
... In the realm of image classification, the latest research endeavors focus on enhancing accuracy through ensemble learning, combining top-performing models. Chouhan et al. [17] achieved an accuracy rate of 96.4% by combining deep CNN models, while Mabrouk et al. [18] merged vision transformer, MobileNetV2, and DenseNet169 for an optimal accuracy of 93.91%. The model we designed introduces improvements by leveraging concepts from GoogLeNet, Xception, and ResNet. ...
A single neural network model cannot capture intricate and diverse features due to its ability to learn only a finite set of patterns from the data. Additionally, training and utilising a single model can be computationally demanding. Experts propose incorporating multiple neural network models to address these constraints to extract complementary attributes. Previous research has highlighted challenges network models face, including difficulties in effectively capturing highly detailed spatial features, redundancy in network structure parameters, and restricted generalisation capabilities. This study introduces an innovative neural network architecture that combines the Xception module with the inverse residue structure to tackle these issues. Considering this, the paper presents a model for detecting pneumonia in X-ray images employing an improved depthwise separable convolutional network. This network architecture integrates the inverse residual structure from the MobileNetV2 model, using the rectified linear unit (ReLU) non-linear activation function throughout the entire network. The experimental results show an impressive recognition rate with a test accuracy of 97.24% on the chest x-ray dataset. This method can extract more profound and abstract image features while mitigating overfitting issues and enhancing the network's generalisation capacity.
... Various sources, including traditional enterprises, social media, machines, and sensor devices, contribute complex data, with up to 80% falling under the category of dark data. This implies that only 10% of the data is deemed useful, while the remaining 90% is classified as dark data [15]. Dark data comprises a substantial portion of big data that is neither beneficial nor usable for significant purposes and is often concealed in cloud and machine storage. ...
________________________________________________________________________________________________________ Abstract: Dark data, or unused information included into routine activities, poses significant hurdles in the era of data-driven decision-making because of its volume and complexity. The goal of this publication is to increase the accuracy of accident prediction by proposing an efficient procedure for dark data extraction and analysis. Data extraction, classifier implementation, and performance evaluation are all done in a methodical manner by using AdaBoost and Random Forest classifiers. According to the results, the Random Forest classifier outperforms the AdaBoost classifier with an accuracy of 89.50%, compared to the former's 78.4%. These results highlight the potential of dark data to yield insightful information by demonstrating how well these classifiers improve accident prediction models. In addition to emphasizing the value of dark data for decision-makers and urban planners looking to improve prediction models and access hidden information, the study offers a methodology for using it. Our research highlights the increasing significance of dark data in enhancing decision-making procedures and forecast precision as data quantities increase.
... Vikash Chouhan et al [8] used a singular deep studying framework for pneumonia detection become proposed, employing switch learning and ensemble type techniques. Five pre-trained CNN models, which includes AlexNet, DenseNet121, ResNet18, InceptionV3, and GoogLeNet, were first-class-tuned on a pneumonia dataset. ...
(1) Background: The COVID-19 pandemic is an unprecedented global challenge, having affected more than 776.79 million people, with over 7.07 million deaths recorded since 2020. The application of Deep Learning (DL) in diagnosing COVID-19 through chest X-rays and computed tomography (CXR and CT) has proven promising. While CNNs have been effective, models such as the Vision Transformer and Swin Transformer have emerged as promising solutions in this field. (2) Methods: This study investigated the performance of models like ResNet50, Vision Transformer, and Swin Transformer. We utilized Bayesian Optimization (BO) in the diagnosis of COVID-19 in CXR and CT based on four distinct datasets: COVID-QU-Ex, HCV-UFPR-COVID-19, HUST-19, and SARS-COV-2 Ct-Scan Dataset. We found that, although all tested models achieved commendable performance metrics, the Swin Transformer stood out. Its unique architecture provided greater generalization power, especially in cross-dataset evaluation (CDE) tasks, where it was trained on one dataset and tested on another. (3) Results: Our approach aligns with state-of-the-art (SOTA) methods, even in complex tasks like CDE. On some datasets, we achieved exceptional performance metrics, with AUC, Accuracy, Precision, Recall, and F1-Score values of 1. (4) Conclusion: Results obtained by the Swin Transformer go beyond what is offered by current SOTA methods and indicate actual feasibility for application in medical diagnostic scenarios. The robustness and generalization power of the Swin Transformer, demonstrated across different datasets, encourage future exploration and adoption of this approach in clinical settings.
Diseases of the airways and the other parts of the lung cause chronic respiratory diseases. The major cause of lung disease is tobacco smoke, along with risk factors such as dust, air pollution, chemicals, and frequent lower respiratory infections during childhood. Early detection of these diseases requires the analysis of medical images, which would aid doctors in providing effective treatment.This paper aims to classify lung X-ray images as benign or malignant and to identify the type of disease, such as Atelectasis, Infiltration, Nodule, and Pneumonia, if the disease is malignant. Machine learning (ML) approaches, combined with a multi-attribute decision-making method called Technique for Order Preference by Similarity to Ideal Solution (TOPSIS), are used to rank different classifiers. Additionally, the deep learning (DL) model Inception v3 is proposed. This method ranks the SVM with RBF as the best classifier among the others used in this approach. Furthermore, the results show that the deep learning model achieves the best accuracy of 97.05%, which is 11.8% higher than the machine learning approach using the same dataset.
Chest X-ray Radiography (CXR) holds significant importance in medical imaging for identifying critical illnesses. While conventional diagnosis relied on visual examination by radiologists, inherent patterns in thoracic diseases have made this method prone to human error, leading to inaccurate diagnoses. Recent developments in the fields of Machine Learning (ML) and Deep Learning (DL) have revolutionized analysis of CXR images, offering efficient and rapid detection methods. Various contemporary DL models, including VGG, ResNet, DenseNet, InceptionNet, EfficientNet, and ensemble learning techniques, have shown promising results in chest disease detection. This paper presents an overview of the latest DL solutions for pneumonia and COVID-19 detection, highlighting recent trends, available datasets, guidelines for implementing DL processes, existing challenges, and prospective research directions. The synthesized findings aim to aid researchers and developers in navigating their work within this domain effectively.
Wheezing, a prevalent respiratory symptom in children, poses diagnostic and management complexities due to its clinical and etiological diversity and the challenge of capturing objective data on pediatric airway inflammation and function. Despite recent advances in understanding pediatric wheezing illnesses, challenges remain in early etiologic diagnosis, phenotype classification, and management. These are mainly attributed to the absence of standardized diagnostic methodologies and effective personalized treatment protocols. The evolution of artificial intelligence (AI) technology introduces new opportunities for managing pediatric wheezing illnesses. The utilization of AI in healthcare has shown considerable promise in disease identification, treatment suggestions, and personalized medicine. Although the application of AI in pediatric wheezing is relatively minimal currently, its utilization in diseases related to pediatric wheezing, specifically in the diagnosis and management of pneumonia and asthma, has seen numerous successful instances. Consequently, this review primarily summarized the application of AI in pediatric wheezing illnesses, and explored the use of AI in pediatric pneumonia and asthma, with an aim to culminate valuable experience from these domains, thereby enriching the application of AI in pediatric wheezing illnesses.
To identify infections caused by the COVID-19 virus, specifically those leading to Pneumonia. The datasets used included images of infected individuals, X-rays of the Chest, and standard non-COVID-19 X-ray images of the chest.
In our research, X-ray chest images were utilized to detect Pneumonia caused by COVID-19. The dataset was collected from the medical database of Johns Hopkins University (USA). To extract detailed features and characteristics from the input images and aid in detecting COVID-19 induced pneumonia cases, a lightweight Stacked Shallow Convolutional Neural Network (CNN) was implemented in the proposed work.
An overall sample size of 2292 input images was considered, with 542 COVID-19 infected images and 1266 non-COVID-19 images used for training. Similarly, for testing, 167 COVID-19 infected images and 317 non-COVID-19 infected images were used. Our proposed model was validated against an established Stacked Shallow CNN to analyze its accuracy. It demonstrated that the proposed framework achieved an accuracy of 98.76%, with each class accuracy of 98.42% for diagnosing standard cases and 99.40% for diagnosing infected COVID-19 cases.
In light of the ongoing research in enhancing the Deep Neural Network model (DNN) to accomplish better accuracy, the CNN architecture presented in this paper introduces a Stacked Shallow Learning approach. This leads to a significant enhancement in both accuracy and efficiency, particularly in computation and response time. Our work outlines a lightweight architecture that is instrumental in expediting the diagnosis of COVID-19 by streamlining response times effectively. Our proposed model also has limitations, such that it cannot hold a large-scale unbalanced dataset and has less scope for pre-processing the data set. Furthermore, the suggested model cannot capture the lung region impacted by the COVID-19 infection.
In this paper, to categorize and detect Pneumonia desease from a collection of chest X-ray picture samples, we propose a deep learning technique based on object detection, convolutional neural networks, and transfer learning. The proposed model is a hybridation of to models the pre-trained model (ResNet -50) and our designed architecture. The Guangzhou Women and Children's Medical Center in Guangzhou, China provided the chest X-ray dataset used in this study. There are 5,000 samples in the data set, with 1,583 healthy samples and 4,273 pneumonia samples). Preprocessing techniques such as Bilateral Equalization was also used to improve accuracy. Due to the imbalance of the data set, we adopted some training techniques to improve the learning process of the samples. This network achieved over 99% accuracy due to the proposed architecture that is based on a combination of two models. The pre-trained ResNet 50 as feature extractor and our designed convolutional neural network (CNN).
Medical Image Analysis (MedIA) has become indispensable in modern healthcare, enhancing clinical diagnostics and personalized treatment. Despite the remarkable advancements supported by deep learning (DL) technologies, their practical deployment faces challenges due to distribution shifts, where models trained on specific datasets underperform across others from varying hospitals, regions, or patient populations. To navigate this issue, researchers have been actively developing strategies to increase the adaptability and robustness of DL models, enabling their effective use in unfamiliar and diverse environments. This paper systematically reviews approaches that apply DL techniques to MedIA systems affected by distribution shifts. Unlike traditional categorizations based on technical specifications, our approach is grounded in the real-world operational constraints faced by healthcare institutions. Specifically, we categorize the existing body of work into Joint Training, Federated Learning, Fine-tuning, and Domain Generalization, with each method tailored to distinct scenarios caused by Data Accessibility, Privacy Concerns, and Collaborative Protocols. This perspective equips researchers with a nuanced understanding of how DL can be strategically deployed to address distribution shifts in MedIA, ensuring diverse and robust medical applications. By delving deeper into these topics, we highlight potential pathways for future research that not only address existing limitations but also push the boundaries of deployable MedIA technologies.
The accurate localization and classification of lung abnormalities from radiological images are important for clinical diagnosis and treatment strategies. However, multilabel classification, wherein medical images are interpreted to point out multiple existing or suspected pathologies, presents practical constraints. Building a highly precise classification model typically requires a huge number of images manually annotated with labels and finding masks that are expensive to acquire in practice. To address this intrinsically weakly supervised learning problem, we present the integration of different features extracted from shallow handcrafted techniques and a pretrained deep CNN model. The model consists of two main approaches: a localization approach that concentrates adaptively on the pathologically abnormal regions utilizing pretrained DenseNet-121 and a classification approach that integrates four types of local and deep features extracted respectively from SIFT, GIST, LBP, and HOG, and convolutional CNN features. We demonstrate that our approaches efficiently leverage interdependencies among target annotations and establish the state of the art classification results of 14 thoracic diseases in comparison with current reference baselines on the publicly available ChestX-ray14 dataset.
Pneumonia is a bacterial, viral, or fungal infection of one or both sides of the lungs that causes lung alveoli to fill up with fluid or pus, which is usually diagnosed with chest x-rays. This work investigates opportunities for applying machine learning solutions for automated detection and localization of pneumonia on chest x-ray images. We propose an ensemble of two convolutional neural networks, namely RetinaNet and Mask R-CNN for pneumonia detection and localization. We validated our solution on a recently released dataset of 26,684 images from Kaggle Pneumonia Detection Challenge and were score among the top 3% of submitted solutions. With 0.793 recall, we developed a reliable solution for automated pneumonia diagnosis and validated it on the largest clinical database publicity available to date. Some of the challenging cases were additionally examined by a team of physicians, who helped us to interpret the obtained results and confirm their practical applicability.
Lung cancer is one of the major causes of cancer-related deaths due to its aggressive nature and delayed detections at advanced stages. Early detection of lung cancer is very important for the survival of an individual, and is a significant challenging problem. Generally, chest radiographs (X-ray) and computed tomography (CT) scans are used initially for the diagnosis of the malignant nodules; however, the possible existence of benign nodules leads to erroneous decisions. At early stages, the benign and the malignant nodules show very close resemblance to each other. In this paper, a novel deep learning-based model with multiple strategies is proposed for the precise diagnosis of the malignant nodules. Due to the recent achievements of deep convolutional neural networks (CNN) in image analysis, we have used two deep three-dimensional (3D) customized mixed link network (CMixNet) architectures for lung nodule detection and classification, respectively. Nodule detections were performed through faster R-CNN on efficiently-learned features from CMixNet and U-Net like encoder–decoder architecture. Classification of the nodules was performed through a gradient boosting machine (GBM) on the learned features from the designed 3D CMixNet structure. To reduce false positives and misdiagnosis results due to different types of errors, the final decision was performed in connection with physiological symptoms and clinical biomarkers. With the advent of the internet of things (IoT) and electro-medical technology, wireless body area networks (WBANs) provide continuous monitoring of patients, which helps in diagnosis of chronic diseases—especially metastatic cancers. The deep learning model for nodules’ detection and classification, combined with clinical factors, helps in the reduction of misdiagnosis and false positive (FP) results in early-stage lung cancer diagnosis. The proposed system was evaluated on LIDC-IDRI datasets in the form of sensitivity (94%) and specificity (91%), and better results were obatined compared to the existing methods.
Currently, lung cancer has one of the highest mortality rates because it is often caught too late. Therefore, early detection is essential to reduce the risk of death. Pulmonary nodules are considered key indicators of primary lung cancer. Developing an efficient and accurate computer-aided diagnosis system for pulmonary nodule detection is an important goal. Typically, a computer-aided diagnosis system for pulmonary nodule detection consists of two parts: candidate nodule extraction and false-positive reduction of candidate nodules. The reduction of false positives (FPs) of candidate nodules remains an important challenge due to morphological characteristics of nodule height changes and similar characteristics to other organs. In this study, we propose a novel multi-scale heterogeneous three-dimensional (3D) convolutional neural network (MSH-CNN) based on chest computed tomography (CT) images. There are three main strategies of the design: (1) using multi-scale 3D nodule blocks with different levels of contextual information as inputs; (2) using two different branches of 3D CNN to extract the expression features; (3) using a set of weights which are determined by back propagation to fuse the expression features produced by step 2. In order to test the performance of the algorithm, we trained and tested on the Lung Nodule Analysis 2016 (LUNA16) dataset, achieving an average competitive performance metric (CPM) score of 0.874 and a sensitivity of 91.7% at two FPs/scan. Moreover, our framework is universal and can be easily extended to other candidate false-positive reduction tasks in 3D object detection, as well as 3D object classification.
Abstract Deep convolutional neural networks have performed remarkably well on many Computer Vision tasks. However, these networks are heavily reliant on big data to avoid overfitting. Overfitting refers to the phenomenon when a network learns a function with very high variance such as to perfectly model the training data. Unfortunately, many application domains do not have access to big data, such as medical image analysis. This survey focuses on Data Augmentation, a data-space solution to the problem of limited data. Data Augmentation encompasses a suite of techniques that enhance the size and quality of training datasets such that better Deep Learning models can be built using them. The image augmentation algorithms discussed in this survey include geometric transformations, color space augmentations, kernel filters, mixing images, random erasing, feature space augmentation, adversarial training, generative adversarial networks, neural style transfer, and meta-learning. The application of augmentation methods based on GANs are heavily covered in this survey. In addition to augmentation techniques, this paper will briefly discuss other characteristics of Data Augmentation such as test-time augmentation, resolution impact, final dataset size, and curriculum learning. This survey will present existing methods for Data Augmentation, promising developments, and meta-level decisions for implementing Data Augmentation. Readers will understand how Data Augmentation can improve the performance of their models and expand limited datasets to take advantage of the capabilities of big data.
Automated diagnosis of tuberculosis (TB) from chest X-Rays (CXR) has been tackled with either hand-crafted algorithms or machine learning approaches such as support vector machines (SVMs) and convolutional neural networks (CNNs). Most deep neural network applied to the task of tuberculosis diagnosis have been adapted from natural image classification. These models have a large number of parameters as well as high hardware requirements, which makes them prone to overfitting and harder to deploy in mobile settings. We propose a simple convolutional neural network optimized for the problem which is faster and more efficient than previous models but preserves their accuracy. Moreover, the visualization capabilities of CNNs have not been fully investigated. We test saliency maps and grad-CAMs as tuberculosis visualization methods, and discuss them from a radiological perspective.
Explainable artificial intelligence (AI) is attracting much interest in medicine. Technically, the problem of explainability is as old as AI itself and classic AI represented comprehensible retraceable approaches. However, their weakness was in dealing with uncertainties of the real world. Through the introduction of probabilistic learning, applications became increasingly successful, but increasingly opaque. Explainable AI deals with the implementation of transparency and traceability of statistical black‐box machine learning methods, particularly deep learning (DL). We argue that there is a need to go beyond explainable AI. To reach a level of explainable medicine we need causability. In the same way that usability encompasses measurements for the quality of use, causability encompasses measurements for the quality of explanations. In this article, we provide some necessary definitions to discriminate between explainability and causability as well as a use‐case of DL interpretation and of human explanation in histopathology. The main contribution of this article is the notion of causability, which is differentiated from explainability in that causability is a property of a person, while explainability is a property of a system
This article is categorized under:
• Fundamental Concepts of Data and Knowledge > Human Centricity and User Interaction
Abstract
Explainable AI.
Detecting anomaly of chest X-ray images by advanced technologies, such as deep learning, is an urgent need to improve the work efficiency and diagnosis accuracy. Fine-tuning existing deep learning networks for medical image processing suffers from over-fitting and low transfer efficiency. To overcome such limitations, we design a hierarchical convolutional neural network (CNN) structure for ChestX-ray14 and propose a new network CXNet-m1, which is much shorter, thinner but more powerful than fine-tuning. we also raise a novel loss function sin-loss, which can learn discriminative information from misclassified and indistinguishable images. Besides, we optimize the convolutional kernels of CXNet-m1 to achieve better classification accuracy. The experimental results show that our light model CXNet-m1 with sin-loss function achieves better accuracy rate, recall rate, F1-score and AUC value. It illustrates that designing a proper CNN is better than fine-tuning deep networks, and the increase of training data is vital to enhance the performance of CNN.
Background
Accurately detecting and examining lung nodules early is key in diagnosing lung cancers and thus one of the best ways to prevent lung cancer deaths. Radiologists spend countless hours detecting small spherical-shaped nodules in computed tomography (CT) images. In addition, even after detecting nodule candidates, a considerable amount of effort and time is required for them to determine whether they are real nodules. The aim of this paper is to introduce a high performance nodule classification method that uses three dimensional deep convolutional neural networks (DCNNs) and an ensemble method to distinguish nodules between non-nodules.
Methods
In this paper, we use a three dimensional deep convolutional neural network (3D DCNN) with shortcut connections and a 3D DCNN with dense connections for lung nodule classification. The shortcut connections and dense connections successfully alleviate the gradient vanishing problem by allowing the gradient to pass quickly and directly. Connections help deep structured networks to obtain general as well as distinctive features of lung nodules. Moreover, we increased the dimension of DCNNs from two to three to capture 3D features. Compared with shallow 3D CNNs used in previous studies, deep 3D CNNs more effectively capture the features of spherical-shaped nodules. In addition, we use an alternative ensemble method called the checkpoint ensemble method to boost performance.
Results
The performance of our nodule classification method is compared with that of the state-of-the-art methods which were used in the LUng Nodule Analysis 2016 Challenge. Our method achieves higher competition performance metric (CPM) scores than the state-of-the-art methods using deep learning. In the experimental setup ESB-ALL, the 3D DCNN with shortcut connections and the 3D DCNN with dense connections using the checkpoint ensemble method achieved the highest CPM score of 0.910.
Conclusion
The result demonstrates that our method of using a 3D DCNN with shortcut connections, a 3D DCNN with dense connections, and the checkpoint ensemble method is effective for capturing 3D features of nodules and distinguishing nodules between non-nodules.
Lung cancer accounts for more than 1.5 million deaths worldwide, and it corresponded to 26% of all deaths due to cancer in 2017. However, lung computer-aided diagnosis systems developed to identify lung cancer at early stages are increasing survival rates. This study explores the performance of deep transfer learning from non-medical images on lung nodule malignancy classification tasks in order to improve such systems. Initially, the 1018 chest computed tomography (CT) examinations and medical annotations from the LIDC/IDRI were processed. Then, several convolutional neural networks (VGG16, VGG19, MobileNet, Xception, InceptionV3, ResNet50, InceptionResNetV2, DenseNet169, DenseNet201, NASNetMobile and NASNetLarge) were built, trained on the ImageNet dataset, converted into feature extractors and applied on the LIDC/IDRI nodule images. Following this, each set of deep features was submitted to 10-fold cross-validations with naive Bayes, multilayer perceptron, support vector machine (SVM), K-nearest neighbors KNN and random forest classifiers. Finally, the evaluation metrics accuracy (ACC), area under the curve (AUC), true positive rate (TPR), precision (PPV) and F1-score of each cross-validation average result were computed and compared. The results showed that the deep feature extractor based on the ResNet50 and the SVM RBF classifier, achieved an AUC metric of 93.1% (the highest value not only among the evaluated combinations, but also among the related works in the literature evaluated), a TPR of 85.38%, an ACC of 88.41%, a PPV of 73.48% and an F1-score of 78.83%. Based on these results, deep transfer learning proves to be a relevant strategy to extract representative features from lung nodule CT images.
Background
Pneumothorax can precipitate a life-threatening emergency due to lung collapse and respiratory or circulatory distress. Pneumothorax is typically detected on chest X-ray; however, treatment is reliant on timely review of radiographs. Since current imaging volumes may result in long worklists of radiographs awaiting review, an automated method of prioritizing X-rays with pneumothorax may reduce time to treatment. Our objective was to create a large human-annotated dataset of chest X-rays containing pneumothorax and to train deep convolutional networks to screen for potentially emergent moderate or large pneumothorax at the time of image acquisition.
Methods and findings
In all, 13,292 frontal chest X-rays (3,107 with pneumothorax) were visually annotated by radiologists. This dataset was used to train and evaluate multiple network architectures. Images showing large- or moderate-sized pneumothorax were considered positive, and those with trace or no pneumothorax were considered negative. Images showing small pneumothorax were excluded from training. Using an internal validation set (n = 1,993), we selected the 2 top-performing models; these models were then evaluated on a held-out internal test set based on area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and positive predictive value (PPV). The final internal test was performed initially on a subset with small pneumothorax excluded (as in training; n = 1,701), then on the full test set (n = 1,990), with small pneumothorax included as positive. External evaluation was performed using the National Institutes of Health (NIH) ChestX-ray14 set, a public dataset labeled for chest pathology based on text reports. All images labeled with pneumothorax were considered positive, because the NIH set does not classify pneumothorax by size. In internal testing, our “high sensitivity model” produced a sensitivity of 0.84 (95% CI 0.78–0.90), specificity of 0.90 (95% CI 0.89–0.92), and AUC of 0.94 for the test subset with small pneumothorax excluded. Our “high specificity model” showed sensitivity of 0.80 (95% CI 0.72–0.86), specificity of 0.97 (95% CI 0.96–0.98), and AUC of 0.96 for this set. PPVs were 0.45 (95% CI 0.39–0.51) and 0.71 (95% CI 0.63–0.77), respectively. Internal testing on the full set showed expected decreased performance (sensitivity 0.55, specificity 0.90, and AUC 0.82 for high sensitivity model and sensitivity 0.45, specificity 0.97, and AUC 0.86 for high specificity model). External testing using the NIH dataset showed some further performance decline (sensitivity 0.28–0.49, specificity 0.85–0.97, and AUC 0.75 for both). Due to labeling differences between internal and external datasets, these findings represent a preliminary step towards external validation.
Conclusions
We trained automated classifiers to detect moderate and large pneumothorax in frontal chest X-rays at high levels of performance on held-out test data. These models may provide a high specificity screening solution to detect moderate or large pneumothorax on images collected when human review might be delayed, such as overnight. They are not intended for unsupervised diagnosis of all pneumothoraces, as many small pneumothoraces (and some larger ones) are not detected by the algorithm. Implementation studies are warranted to develop appropriate, effective clinician alerts for the potentially critical finding of pneumothorax, and to assess their impact on reducing time to treatment.
Background
Chest radiograph interpretation is critical for the detection of thoracic diseases, including tuberculosis and lung cancer, which affect millions of people worldwide each year. This time-consuming task typically requires expert radiologists to read the images, leading to fatigue-based diagnostic error and lack of diagnostic expertise in areas of the world where radiologists are not available. Recently, deep learning approaches have been able to achieve expert-level performance in medical image interpretation tasks, powered by large network architectures and fueled by the emergence of large labeled datasets. The purpose of this study is to investigate the performance of a deep learning algorithm on the detection of pathologies in chest radiographs compared with practicing radiologists.
Methods and findings
We developed CheXNeXt, a convolutional neural network to concurrently detect the presence of 14 different pathologies, including pneumonia, pleural effusion, pulmonary masses, and nodules in frontal-view chest radiographs. CheXNeXt was trained and internally validated on the ChestX-ray8 dataset, with a held-out validation set consisting of 420 images, sampled to contain at least 50 cases of each of the original pathology labels. On this validation set, the majority vote of a panel of 3 board-certified cardiothoracic specialist radiologists served as reference standard. We compared CheXNeXt’s discriminative performance on the validation set to the performance of 9 radiologists using the area under the receiver operating characteristic curve (AUC). The radiologists included 6 board-certified radiologists (average experience 12 years, range 4–28 years) and 3 senior radiology residents, from 3 academic institutions. We found that CheXNeXt achieved radiologist-level performance on 11 pathologies and did not achieve radiologist-level performance on 3 pathologies. The radiologists achieved statistically significantly higher AUC performance on cardiomegaly, emphysema, and hiatal hernia, with AUCs of 0.888 (95% confidence interval [CI] 0.863–0.910), 0.911 (95% CI 0.866–0.947), and 0.985 (95% CI 0.974–0.991), respectively, whereas CheXNeXt’s AUCs were 0.831 (95% CI 0.790–0.870), 0.704 (95% CI 0.567–0.833), and 0.851 (95% CI 0.785–0.909), respectively. CheXNeXt performed better than radiologists in detecting atelectasis, with an AUC of 0.862 (95% CI 0.825–0.895), statistically significantly higher than radiologists' AUC of 0.808 (95% CI 0.777–0.838); there were no statistically significant differences in AUCs for the other 10 pathologies. The average time to interpret the 420 images in the validation set was substantially longer for the radiologists (240 minutes) than for CheXNeXt (1.5 minutes). The main limitations of our study are that neither CheXNeXt nor the radiologists were permitted to use patient history or review prior examinations and that evaluation was limited to a dataset from a single institution.
Conclusions
In this study, we developed and validated a deep learning algorithm that classified clinically important abnormalities in chest radiographs at a performance level comparable to practicing radiologists. Once tested prospectively in clinical settings, the algorithm could have the potential to expand patient access to chest radiograph diagnostics.
In this article, the authors aim to maximally utilize multimodality neuroimaging and genetic data for identifying Alzheimer's disease (AD) and its prodromal status, Mild Cognitive Impairment (MCI), from normal aging subjects. Multimodality neuroimaging data such as MRI and PET provide valuable insights into brain abnormalities, while genetic data such as single nucleotide polymorphism (SNP) provide information about a patient's AD risk factors. When these data are used together, the accuracy of AD diagnosis may be improved. However, these data are heterogeneous (e.g., with different data distributions), and have different number of samples (e.g., with far less number of PET samples than the number of MRI or SNPs). Thus, learning an effective model using these data is challenging. To this end, we present a novel three‐stage deep feature learning and fusion framework, where deep neural network is trained stage‐wise. Each stage of the network learns feature representations for different combinations of modalities, via effective training using the maximum number of available samples. Specifically, in the first stage, we learn latent representations (i.e., high‐level features) for each modality independently, so that the heterogeneity among modalities can be partially addressed, and high‐level features from different modalities can be combined in the next stage. In the second stage, we learn joint latent features for each pair of modality combination by using the high‐level features learned from the first stage. In the third stage, we learn the diagnostic labels by fusing the learned joint latent features from the second stage. To further increase the number of samples during training, we also use data at multiple scanning time points for each training subject in the dataset. We evaluate the proposed framework using Alzheimer's disease neuroimaging initiative (ADNI) dataset for AD diagnosis, and the experimental results show that the proposed framework outperforms other state‐of‐the‐art methods.
In this review the application of deep learning for medical diagnosis is addressed. A thorough analysis of various scientific articles in the domain of deep neural networks application in the medical field has been conducted. More than 300 research articles were obtained, and after several selection steps, 46 articles were presented in more detail. The results indicate that convolutional neural networks (CNN) are the most widely represented when it comes to deep learning and medical image analysis. Furthermore, based on the findings of this article, it can be noted that the application of deep learning technology is widespread, but the majority of applications are focused on bioinformatics, medical diagnosis and other similar fields.
The extraction of features from the fully connected layer of a convolutional neural network (CNN) model is widely used for image representation. However, the features obtained by the convolutional layers are seldom investigated due to their high dimensionality and lack of global representation. In this study, we explore the uses of local description and feature encoding for deeply convolutional features. Given an input image, the image pyramid is constructed, and different pretrained CNNs are applied to each image scale to extract convolutional features. Deeply local descriptors can be obtained by concatenating the convolutional features in each spatial position. Hellinger kernel and principal component analysis (PCA) are introduced to improve the distinguishable capabilities of the deeply local descriptors. The Hellinger kernel causes the distance measure to be sensitive to small feature values, and the PCA helps reduce feature redundancy. In addition, two aggregate strategies are proposed to form global image representations from the deeply local descriptors. The first strategy aggregates the descriptors of different CNNs by Fisher encoding, and the second strategy concatenates the Fisher vectors of different CNNs. Experiments on two remote sensing image datasets illustrate that the Hellinger kernel, PCA, and two aggregate strategies improve classification performance. Moreover, the deeply local descriptors outperform the features extracted from fully connected layers.
Recent work has shown that convolutional networks can be substantially deeper, more accurate and efficient to train if they contain shorter connections between layers close to the input and those close to the output. In this paper we embrace this observation and introduce the Dense Convolutional Network (DenseNet), where each layer is directly connected to every other layer in a feed-forward fashion. Whereas traditional convolutional networks with L layers have L connections, one between each layer and its subsequent layer (treating the input as layer 0), our network has L(L+1)/2 direct connections. For each layer, the feature maps of all preceding layers are treated as separate inputs whereas its own feature maps are passed on as inputs to all subsequent layers. Our proposed connectivity pattern has several compelling advantages: it alleviates the vanishing gradient problem and strengthens feature propagation; despite the increase in connections, it encourages feature reuse and leads to a substantial reduction of parameters; its models tend to generalize surprisingly well. We evaluate our proposed architecture on five highly competitive object recognition benchmark tasks. The DenseNet obtains significant improvements over the state-of-the-art on all five of them (e.g., yielding 3.74% test error on CIFAR-10, 19.25% on CIFAR-100 and 1.59% on SVHN).
In recent years, many publications showed that convolutional neural network based features can have a superior performance to engineered features. However, not much effort was taken so far to extract local features efficiently for a whole image. In this paper, we present an approach to compute patch-based local feature descriptors efficiently in presence of pooling and striding layers for whole images at once. Our approach is generic and can be applied to nearly all existing network architectures. This includes networks for all local feature extraction tasks like camera calibration, Patchmatching, optical flow estimation and stereo matching. In addition, our approach can be applied to other patch-based approaches like sliding window object detection and recognition. We complete our paper with a speed benchmark of popular CNN based feature extraction approaches applied on a whole image, with and without our speedup, and example code (for Torch) that shows how an arbitrary CNN architecture can be easily converted by our approach.
The term Deep Learning or Deep Neural Network refers to Artificial Neural Networks (ANN) with multi layers . Over the last few decades, it has been considered to be one of the most powerful tools, and has become very popular in the literature as it is able to handle a huge amount of data. The interest in having deeper hidden layers has recently begun to surpass classical methods performance in different fields; especially in pattern recognition. One of the most popular deep neural networks is the Convolutional Neural Network (CNN). It take this name from mathematical linear operation between matrixes called convolution. CNN have multiple layers; including convolutional layer, non-linearity layer, pooling layer and fully-connected layer. The convolutional and fully- connected layers have parameters but pooling and non-linearity layers don't have parameters. The CNN has an excellent performance in machine learning problems. Specially the applications that deal with image data, such as largest image classification data set (Image Net), computer vision, and in natural language processing (NLP) and the results achieved were very amazing . In this paper we will explain and define all the elements and important issues related to CNN, and how these elements work. In addition, we will also state the parameters that effect CNN efficiency. This paper assumes that the readers have adequate knowledge about both machine learning and artificial neural network.
The chest X-ray is one of the most commonly accessible radiological examinations for screening and diagnosis of many lung diseases. A tremendous number of X-ray imaging studies accompanied by radiological reports are accumulated and stored in many modern hospitals' Picture Archiving and Communication Systems (PACS). On the other side, it is still an open question how this type of hospital-size knowledge database containing invaluable imaging informatics (i.e., loosely labeled) can be used to facilitate the data-hungry deep learning paradigms in building truly large-scale high precision computer-aided diagnosis (CAD) systems. In this paper, we present a new chest X-ray database, namely " ChestX-ray8 " , which comprises 108,948 frontal-view X-ray images of 32,717 unique patients with the text-mined eight disease image labels (where each image can have multi-labels), from the associated radiological reports using natural language processing. Importantly, we demonstrate that these commonly occurring thoracic diseases can be detected and even spatially-located via a unified weakly-supervised multi-label image classification and disease lo-calization framework, which is validated using our proposed dataset. Although the initial quantitative results are promising as reported, deep convolutional neural network based " reading chest X-rays " (i.e., recognizing and locating the common disease patterns trained with only image-level labels) remains a strenuous task for fully-automated high precision CAD systems.
The chest X-ray is one of the most commonly accessible radiological examinations for screening and diagnosis of many lung diseases. A tremendous number of X-ray imaging studies accompanied by radiological reports are accumulated and stored in many modern hospitals' Picture Archiving and Communication Systems (PACS). On the other side, it is still an open question how this type of hospital-size knowledge database containing invaluable imaging informatics (i.e., loosely labeled) can be used to facilitate the data-hungry deep learning paradigms in building truly large-scale high precision computer-aided diagnosis (CAD) systems. In this paper, we present a new chest X-ray database, namely "ChestX-ray8", which comprises 108,948 frontal-view X-ray images of 32,717 unique patients with the text-mined eight disease image labels (where each image can have multi-labels), from the associated radiological reports using natural language processing. Importantly, we demonstrate that these commonly occurring thoracic diseases can be detected and even spatially-located via a unified weakly-supervised multi-label image classification and disease localization framework, which is validated using our proposed dataset. Although the initial quantitative results are promising as reported, deep convolutional neural network based "reading chest X-rays" (i.e., recognizing and locating the common disease patterns trained with only image-level labels) remains a strenuous task for fully-automated high precision CAD systems.
The ability to automatically learn task specific feature representations has led to a huge success of deep learning methods. When large training data is scarce, such as in medical imaging problems, transfer learning has been very effective. In this paper, we systematically investigate the process of transferring a Convolutional Neural Network, trained on ImageNet images to perform image classification, to kidney detection problem in ultrasound images. We study how the detection performance depends on the extent of transfer. We show that a transferred and tuned CNN can outperform a state-of-the-art feature engineered pipeline and a hybridization of these two techniques achieves 20\% higher performance. We also investigate how the evolution of intermediate response images from our network. Finally, we compare these responses to state-of-the-art image processing filters in order to gain greater insight into how transfer learning is able to effectively manage widely varying imaging regimes.
Deep learning algorithms, in particular convolutional networks, have rapidly become a methodology of choice for analyzing medical images. This paper reviews the major deep learning concepts pertinent to medical image analysis and summarizes over 300 contributions to the field, most of which appeared in the last year. We survey the use of deep learning for image classification, object detection, segmentation, registration, and other tasks and provide concise overviews of studies per application area. Open challenges and directions for future research are discussed.
This book constitutes the refereed proceedings of two workshops held at the 19th International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2016, in Athens, Greece, in October 2016: the First Workshop on Large-Scale Annotation of Biomedical Data and Expert Label Synthesis, LABELS 2016, and the Second International Workshop on Deep Learning in Medical Image Analysis, DLMIA 2016. The 28 revised regular papers presented in this book were carefully reviewed and selected from a total of 52 submissions. The 7 papers selected for LABELS deal with topics from the following fields: crowd-sourcing methods; active learning; transfer learning; semi-supervised learning; and modeling of label uncertainty.The 21 papers selected for DLMIA span a wide range of topics such as image description; medical imaging-based diagnosis; medical signal-based diagnosis; medical image reconstruction and model selection using deep learning techniques; meta-heuristic techniques for fine-tuning parameter in deep learning-based architectures; and applications based on deep learning techniques.
Recent work has shown that convolutional networks can be substantially deeper, more accurate, and efficient to train if they contain shorter connections between layers close to the input and those close to the output. In this paper, we embrace this observation and introduce the Dense Convolutional Network (DenseNet), which connects each layer to every other layer in a feed-forward fashion. Whereas traditional convolutional networks with L layers have L connections - one between each layer and its subsequent layer - our network has L(L+1)/2 direct connections. For each layer, the feature-maps of all preceding layers are used as inputs, and its own feature-maps are used as inputs into all subsequent layers. DenseNets have several compelling advantages: they alleviate the vanishing-gradient problem, strengthen feature propagation, encourage feature reuse, and substantially reduce the number of parameters. We evaluate our proposed architecture on four highly competitive object recognition benchmark tasks (CIFAR-10, CIFAR-100, SVHN, and ImageNet). DenseNets obtain significant improvements over the state-of-the-art on most of them, whilst requiring less memory and computation to achieve high performance. Code and models are available at https://github.com/liuzhuang13/DenseNet.
Background and objective:
Chest X-ray (CXR) is one of the most used imaging techniques for detection and diagnosis of pulmonary diseases. A critical component in any computer-aided system, for either detection or diagnosis in digital CXR, is the automatic segmentation of the lung field. One of the main challenges inherent to this task is to include in the segmentation the lung regions overlapped by dense abnormalities, also known as opacities, which can be caused by diseases such as tuberculosis and pneumonia. This specific task is difficult because opacities frequently reach high intensity values which can be incorrectly interpreted by an automatic method as the lung boundary, and as a consequence, this creates a challenge in the segmentation process, because the chances of incomplete segmentations are increased considerably. The purpose of this work is to propose a method for automatic segmentation of lungs in CXR that addresses this problem by reconstructing the lung regions "lost" due to pulmonary abnormalities.
Methods:
The proposed method, which features two deep convolutional neural network models, consists of four steps main steps: (1) image acquisition, (2) initial segmentation, (3) reconstruction and (4) final segmentation.
Results:
The proposed method was experimented on 138 Chest X-ray images from Montgomery County's Tuberculosis Control Program, and has achieved as best result an average sensitivity of 97.54%, an average specificity of 96.79%, an average accuracy of 96.97%, an average Dice coefficient of 94%, and an average Jaccard index of 88.07%.
Conclusions:
We demonstrate in our lung segmentation method that the problem of dense abnormalities in Chest X-rays can be efficiently addressed by performing a reconstruction step based on a deep convolutional neural network model.
Background and objective:
In most patients presenting with respiratory symptoms, the findings of chest radiography play a key role in the diagnosis, management, and follow-up of the disease. Consolidation is a common term in radiology, which indicates focally increased lung density. When the alveolar structures become filled with pus, fluid, blood cells or protein subsequent to a pulmonary pathological process, it may result in different types of lung opacity in chest radiograph. This study aims at detecting consolidations in chest x-ray radiographs, with a certain precision, using artificial intelligence and especially Deep Convolutional Neural Networks to assist radiologist for better diagnosis.
Methods:
Medical image datasets usually are relatively small to be used for training a Deep Convolutional Neural Network (DCNN), so transfer learning technique with well-known DCNNs pre-trained with ImageNet dataset are used to improve the accuracy of the models. ImageNet feature space is different from medical images and in the other side, the well-known DCNNs are designed to achieve the best performance on ImageNet. Therefore, they cannot show their best performance on medical images. To overcome this problem, we designed a problem-based architecture which preserves the information of images for detecting consolidation in Pediatric Chest X-ray dataset. We proposed a three-step pre-processing approach to enhance generalization of the models. To demonstrate the correctness of numerical results, an occlusion test is applied to visualize outputs of the model and localize the detected appropriate area. A different dataset as an extra validation is used in order to investigate the generalization of the proposed model.
Results:
The best accuracy to detect consolidation is 94.67% obtained by our problem based architecture for the understudy dataset which outperforms the previous works and the other architectures.
Conclusions:
The designed models can be employed as computer aided diagnosis tools in real practice. We critically discussed the datasets and the previous works based on them and show that without some considerations the results of them may be misleading. We believe, the output of AI should be only interpreted as focal consolidation. The clinical significance of the finding can not be interpreted without integration of clinical data.
Activation functions lie at the core of deep neural networks allowing them to learn arbitrarily complex mappings. Without any activation, a neural network learn will only be able to learn a linear relation between input and the desired output. The chapter introduces the reader to why activation functions are useful and their immense importance in making deep learning successful. A detailed survey of several existing activation functions is provided in this chapter covering their functional forms, original motivations, merits as well as demerits. The chapter also discusses the domain of learnable activation functions and proposes a novel activation ‘SLAF’ whose shape is learned during the training of a neural network. A working model for SLAF is provided and its performance is experimentally shown on XOR and MNIST classification tasks.
Diabetic retinopathy (DR) results in vision loss if not treated early. A computer-aided diagnosis (CAD) system based on retinal fundus images is an efficient and effective method for early DR diagnosis and assisting experts. A computer-aided diagnosis (CAD) system involves various stages like detection, segmentation and classification of lesions in fundus images. Many traditional machine-learning (ML) techniques based on hand-engineered features have been introduced. The recent emergence of deep learning (DL) and its decisive victory over traditional ML methods for various applications motivated the researchers to employ it for DR diagnosis, and many deep-learning-based methods have been introduced. In this paper, we review these methods, highlighting their pros and cons. In addition, we point out the challenges to be addressed in designing and learning about efficient, effective and robust deep-learning algorithms for various problems in DR diagnosis and draw attention to directions for future research.
Classification of benign-malignant lung nodules on chest CT is the most critical step in the early detection of lung cancer and prolongation of patient survival. Despite their success in image classification, deep convolutional neural networks (DCNNs) always require a large number of labeled training data, which are not available for most medical image analysis applications due to the work required in image acquisition and particularly image annotation. In this paper, we propose a semi-supervised adversarial classification (SSAC) model that can be trained by using both labeled and unlabeled data for benign-malignant lung nodule classification. This model consists of an adversarial autoencoder-based unsupervised reconstruction network R, a supervised classification network C, and learnable transition layers that enable the adaption of the image representation ability learned by R to C. The SSAC model has been extended to the multi-view knowledge-based collaborative learning, aiming to employ three SSACs to characterize each nodule's overall appearance, heterogeneity in shape and texture, respectively, and to perform such characterization on nine planar views. The MK-SSAC model has been evaluated on the benchmark LIDC-IDRI dataset and achieves an accuracy of 92.53% and an AUC of 95.81%, which are superior to the performance of other lung nodule classification and semi-supervised learning approaches.
Background and objective:
Computer aided diagnosis systems based on deep learning and medical imaging is increasingly becoming research hotspots. At the moment, the classical convolutional neural network generates classification results by hierarchically abstracting the original image. These abstract features are less sensitive to the position and orientation of the object, and this lack of spatial information limits the further improvement of image classification accuracy. Therefore, how to develop a suitable neural network framework and training strategy in practical clinical applications to avoid this problem is a topic that researchers need to continue to explore.
Methods:
We propose a deep learning framework that combines residual thought and dilated convolution to diagnose and detect childhood pneumonia. Specifically, based on an understanding of the nature of the child pneumonia image classification task, the proposed method uses the residual structure to overcome the over-fitting and the degradation problems of the depth model, and utilizes dilated convolution to overcome the problem of loss of feature space information caused by the increment in depth of the model. Furthermore, in order to overcome the problem of difficulty in training model due to insufficient data and the negative impact of the introduction of structured noise on the performance of the model, we use the model parameters learned on large-scale datasets in the same field to initialize our model through transfer learning.
Results:
Our proposed method has been evaluated for extracting texture features associated with pneumonia and for accurately identifying the performance of areas of the image that best indicate pneumonia. The experimental results of the test dataset show that the recall rate of the method on children pneumonia classification task is 96.7%, and the f1-score is 92.7%. Compared with the prior art methods, this approach can effectively solve the problem of low image resolution and partial occlusion of the inflammatory area in children chest X-ray images.
Conclusions:
The novel framework focuses on the application of advanced classification that directly performs lesion characterization, and has high reliability in the classification task of children pneumonia.
The rich collection of annotated datasets piloted the robustness of deep learning techniques to effectuate the implementation of diverse medical imaging tasks. Over 15% of deaths include children under age five are caused by pneumonia globally. In this study, we describe our deep learning based approach for the identification and localization of pneumonia in Chest X-rays (CXRs) images. Researchers usually employ CXRs for the diagnostic imaging study. Several factors such as positioning of the patient and depth of inspiration can change the appearance of the chest X-ray, complicating interpretation further. Our identification model (https://github.com/amitkumarj441/identify_pneumonia) is based on Mask-RCNN, a deep neural network which incorporates global and local features for pixel-wise segmentation. Our approach achieves robustness through critical modifications of the training process and a novel post-processing step which merges bounding boxes from multiple models. The proposed identification model achieves better performances evaluated on chest radiograph dataset which depict potential pneumonia causes.
Remarkable progress has been made in image classification and segmentation, due to the recent study of deep convolutional neural networks (CNNs). To solve the similar problem of diagnostic lung nodule detection in low-dose computed tomography (CT) scans, we propose a new Computer-Aided Detection (CAD) system using CNNs and CT image segmentation techniques. Unlike former studies focusing on the classification of malignant nodule types or relying on prior image processing, in this work, we put raw CT image patches directly in CNNs to reduce the complexity of the system. Specifically, we split each CT image into several patches, which are divided into 6 types consisting of 3 nodule types and 3 non-nodule types. We compare the performance of ResNet with different CNNs architectures on CT images from a publicly available dataset named the Lung Image Database Consortium and Image Database Resource Initiative (LIDC-IDRI). Results show that our best model reaches a high detection sensitivity of 92.8% with 8 false positives per scan (FPs/scan). Compared with related work, our work obtains a state-of-the-art effect.
Background and objective: The X-ray screening is one of the most popular methodologies in detection of respiratory system diseases. Chest organs are screened on the film or digital file which go to the doctor for evaluation. However, the analysis of x-ray images requires much experience and time. Clinical decision support is very important for medical examinations. The use of Computational Intelligence can simulate the evaluation and decision processes of a medical expert. We propose a method to provide a decision support for the doctor in order to help to consult each case faster and more precisely. Methods: We use image descriptors based on the spatial distribution of Hue, Saturation and Brightness values in x-ray images, and a neural network co-working with heuristic algorithms (Moth-Flame, Ant Lion) to detect degenerated lung tissues in x-ray image. The neural network evaluates the image and if the possibility of a respiratory disease is detected, the heuristic method identifies the degenerated tissues in the x-ray image in detail based on the use of the proposed fitness function. Results: The average accuracy is 79.06% in pre-detection stage, similarly the sensitivity and the specificity averaged for three pre-classified diseases are 84.22% and 66.7%, respectively. The misclassification errors are 3.23% for false positives and 3.76% for false negatives. Conclusions: The proposed neuro-heuristic approach addresses small changes in the structure of lung tissues, which appear in pneumonia, sarcoidosis or cancer and some consequences that may appear after the treatment. The results show high potential of the newly proposed method. Additionally, the method is flexible and has low computational burden.
Due to recent advances in artificial intelligence, there is renewed interest in automating interpretation of imaging tests. Chest radiographs are particularly interesting due to many factors: relatively inexpensive equipment, importance to public health, commonly performed throughout the world, and deceptively complex taking years to master. This article presents a brief introduction to artificial intelligence, reviews the progress to date in chest radiograph interpretation, and provides a snapshot of the available datasets and algorithms available to chest radiograph researchers. Finally, the limitations of artificial intelligence with respect to interpretation of imaging studies are discussed.
Deep 2D convolutional neural networks (CNNs) have been remarkably successful in producing record-breaking results in a variety of computer vision tasks. It is possible to extend CNNs to three dimensions using 3D kernels to make them suitable for volumetric medical imaging data such as CT or MRI, but this increases the processing time as well as the required number of training samples (due to the higher number of parameters that need to be learned). In this work, we address both of these issues for a 3D CNN implementation through the development of a two-stage computer-aided detection system for automatic detection of pulmonary nodules. The first stage consists of a 3D fully convolutional network (FCN) for fast screening and generation of candidate suspicious regions. The second stage consists of an ensemble of 3D CNNs trained using extensive transformations applied to both the positive and negative patches to augment the training set. To enable the second stage classifiers to learn differently, they are trained on false positive patches obtained from the screening model using different thresholds on their associated scores as well as different augmentation types. The networks in the second stage are averaged together to produce the final classification score for each candidate patch. Using this procedure, our overall nodule detection system called DeepMed is fast and can achieve 91% sensitivity at 2 false positives per scan on cases from the LIDC dataset.
Computer Aided Diagnosis (CAD) systems can support physicians in classifying different kinds of breast cancer, liver cancer and blood tumours also revealed by images acquired via Computer Tomography, Magnetic Resonance, and Blood Smear systems. In this regard, this survey focuses on papers dealing with the description of existing CAD frameworks for the classification of the three mentioned diseases, by detailing existing CAD workflows based on the same steps for supporting the diagnosis of these tumours. In detail, after an appropriate acquisition of the images, the fundamental steps carried out by a CAD framework can be identified as image segmentation, feature extraction and classification. In particular, in this work, specific CAD frameworks are considered, where the task of feature extraction is performed by using both traditional handcrafted strategies and Convolutional Neural Networks-based innovative methodologies, whereas the final supervised pattern classification is based on neural/non-neural machine learning methods. The cited methodology is focused on sharing and reviewing an amount of specific works. Then, the performance of three selected case studies are carefully reported, designed with the aim of showing how final outcomes can vary according to different choices in each step of the adopted workflow. More in detail, these case studies concern with breast images acquired by Tomosynthesis and Magnetic Resonance, hepatocellular carcinoma images acquired by Computed Tomography and enhanced by a triphasic protocol with a contrast medium, peripheral blood smear images for cellular blood tumours and are used to compare their performance.
Objective:
A novel computer-aided detection (CAD) scheme for lung nodule detection using a 3D deep convolutional neural network combined with a multi-scale prediction strategy is proposed to assist radiologists by providing a second opinion on accurate lung nodule detection, which is a crucial step in early diagnosis of lung cancer.
Method:
A 3D deep convolutional neural network (CNN) with multi-scale prediction was used to detect lung nodules after the lungs were segmented from chest CT scans, with a comprehensive method utilized. Compared with a 2D CNN, a 3D CNN can utilize richer spatial 3D contextual information and generate more discriminative features after being trained with 3D samples to fully represent lung nodules. Furthermore, a multi-scale lung nodule prediction strategy, including multi-scale cube prediction and cube clustering, is also proposed to detect extremely small nodules.
Result:
The proposed method was evaluated on 888 thin-slice scans with 1186 nodules in the LUNA16 database. All results were obtained via 10-fold cross-validation. Three options of the proposed scheme are provided for selection according to the actual needs. The sensitivity of the proposed scheme with the primary option reached 87.94% and 92.93% at one and four false positives per scan, respectively. Meanwhile, the competition performance metric (CPM) score is very satisfying (0.7967).
Conclusion:
The experimental results demonstrate the outstanding detection performance of the proposed nodule detection scheme. In addition, the proposed scheme can be extended to other medical image recognition fields.
Purpose To develop and validate a deep learning-based automatic detection algorithm (DLAD) for malignant pulmonary nodules on chest radiographs and to compare its performance with physicians including thoracic radiologists. Materials and Methods For this retrospective study, DLAD was developed by using 43 292 chest radiographs (normal radiograph-to-nodule radiograph ratio, 34 067:9225) in 34 676 patients (healthy-to-nodule ratio, 30 784:3892; 19 230 men [mean age, 52.8 years; age range, 18-99 years]; 15 446 women [mean age, 52.3 years; age range, 18-98 years]) obtained between 2010 and 2015, which were labeled and partially annotated by 13 board-certified radiologists, in a convolutional neural network. Radiograph classification and nodule detection performances of DLAD were validated by using one internal and four external data sets from three South Korean hospitals and one U.S. hospital. For internal and external validation, radiograph classification and nodule detection performances of DLAD were evaluated by using the area under the receiver operating characteristic curve (AUROC) and jackknife alternative free-response receiver-operating characteristic (JAFROC) figure of merit (FOM), respectively. An observer performance test involving 18 physicians, including nine board-certified radiologists, was conducted by using one of the four external validation data sets. Performances of DLAD, physicians, and physicians assisted with DLAD were evaluated and compared. Results According to one internal and four external validation data sets, radiograph classification and nodule detection performances of DLAD were a range of 0.92-0.99 (AUROC) and 0.831-0.924 (JAFROC FOM), respectively. DLAD showed a higher AUROC and JAFROC FOM at the observer performance test than 17 of 18 and 15 of 18 physicians, respectively (P < .05), and all physicians showed improved nodule detection performances with DLAD (mean JAFROC FOM improvement, 0.043; range, 0.006-0.190; P < .05). Conclusion This deep learning-based automatic detection algorithm outperformed physicians in radiograph classification and nodule detection performance for malignant pulmonary nodules on chest radiographs, and it enhanced physicians' performances when used as a second reader. © RSNA, 2018 Online supplemental material is available for this article.
The paper presents a computer vision based system, which performs real time path finding for visually impaired or blind people. The semantic segmentation of camera images is performed using deep convolutional neural network (CNN), which able to recognize patterns across image feature space. Out of three different CNN architectures (AlexNet, GoogLeNet and VGG) analysed, the fully connected VGG16 neural network is shown to perform best in the semantic segmentation task. The algorithm for extracting and finding paths, obstacles and path boundaries is presented. The experiments performed using own dataset (300 images extracted from two hours of video recording walking in outdoors environment) show that the developed system is able to find paths, path objects and path boundaries with an accuracy of 96.1 ± 2.6%.
Aim:
To develop a machine learning-based model for the binary classification of chest radiography abnormalities, to serve as a retrospective tool in guiding clinician reporting prioritisation.
Materials and methods:
The open-source machine learning library, Tensorflow, was used to retrain a final layer of the deep convolutional neural network, Inception, to perform binary normality classification on two, anonymised, public image datasets. Re-training was performed on 47,644 images using commodity hardware, with validation testing on 5,505 previously unseen radiographs. Confusion matrix analysis was performed to derive diagnostic utility metrics.
Results:
A final model accuracy of 94.6% (95% confidence interval [CI]: 94.3-94.7%) based on an unseen testing subset (n=5,505) was obtained, yielding a sensitivity of 94.6% (95% CI: 94.4-94.7%) and a specificity of 93.4% (95% CI: 87.2-96.9%) with a positive predictive value (PPV) of 99.8% (95% CI: 99.7-99.9%) and area under the curve (AUC) of 0.98 (95% CI: 0.97-0.99).
Conclusion:
This study demonstrates the application of a machine learning-based approach to classify chest radiographs as normal or abnormal. Its application to real-world datasets may be warranted in optimising clinician workload.
More than 50% of cancer patients are treated with radiotherapy, either exclusively or in combination with other methods. The planning and delivery of radiotherapy treatment is a complex process, but can now be greatly facilitated by artificial intelligence technology. Deep learning is the fastest-growing field in artificial intelligence and has been successfully used in recent years in many domains, including medicine. In this article, we first explain the concept of deep learning, addressing it in the broader context of machine learning. The most common network architectures are presented, with a more specific focus on convolutional neural networks. We then present a review of the published works on deep learning methods that can be applied to radiotherapy, which are classified into seven categories related to the patient workflow, and can provide some insights of potential future applications. We have attempted to make this paper accessible to both radiotherapy and deep learning communities, and hope that it will inspire new collaborations between these two communities to develop dedicated radiotherapy applications.
Background and objective:
In medical examinations doctors use various techniques in order to provide to the patients an accurate analysis of their actual state of health. One of the commonly used methodologies is the x-ray screening. This examination very often help to diagnose some diseases of chest organs. The most frequent cause of wrong diagnosis lie in the radiologist's difficulty in interpreting the presence of lungs carcinoma in chest X-ray. In such circumstances, an automated approach could be highly advantageous as it provides important help in medical diagnosis.
Methods:
In this paper we propose a new classification method of the lung carcinomas. This method start with the localization and extraction of the lung nodules by computing, for each pixel of the original image, the local variance obtaining an output image (variance image) with the same size of the original image. In the variance image we find the local maxima and then by using the locations of these maxima in the original image we found the contours of the possible nodules in lung tissues. However after this segmentation stage we find many false nodules. Therefore to discriminate the true ones we use a probabilistic neural network as classifier.
Results:
The performance of our approach is 92% of correct classifications, while the sensitivity is 95% and the specificity is 89.7%. The misclassification errors are due to the fact that network confuses false nodules with the true ones (6%) and true nodules with the false ones (2%).
Conclusions:
Several researchers have proposed automated algorithms to detect and classify pulmonary nodules but these methods fail to detect low-contrast nodules and have a high computational complexity, in contrast our method is relatively simple but at the same time provides good results and can detect low-contrast nodules. Furthermore, in this paper is presented a new algorithm for training the PNN neural networks that allows to obtain PNNs with many fewer neurons compared to the neural networks obtained by using the training algorithms present in the literature. So considerably lowering the computational burden of the trained network and at same time keeping the same performances.
The implementation of clinical-decision support algorithms for medical imaging faces challenges with reliability and interpretability. Here, we establish a diagnostic tool based on a deep-learning framework for the screening of patients with common treatable blinding retinal diseases. Our framework utilizes transfer learning, which trains a neural network with a fraction of the data of conventional approaches. Applying this approach to a dataset of optical coherence tomography images, we demonstrate performance comparable to that of human experts in classifying age-related macular degeneration and diabetic macular edema. We also provide a more transparent and interpretable diagnosis by highlighting the regions recognized by the neural network. We further demonstrate the general applicability of our AI system for diagnosis of pediatric pneumonia using chest X-ray images. This tool may ultimately aid in expediting the diagnosis and referral of these treatable conditions, thereby facilitating earlier treatment, resulting in improved clinical outcomes.
Video Abstract
Download video (30MB)Help with mp4 files
In this paper, we propose a deep learning approach for image registration by predicting deformation from image appearance. Since obtaining ground-truth deformation fields for training can be challenging, we design a fully convolutional network that is subject to dual-guidance: (1) Coarse guidance using deformation fields obtained by an existing registration method; and (2) Fine guidance using image similarity. The latter guidance helps avoid overly relying on the supervision from the training deformation fields, which could be inaccurate. For effective training, we further improve the deep convolutional network with gap filling, hierarchical loss, and multi-source strategies. Experiments on a variety of datasets show promising registration accuracy and efficiency compared with state-of-the-art methods.
Community acquired pneumonia remains a common cause of morbidity and mortality. Usually, the causal organism is not identified and treatment remains empiric. Recent computed tomography and magnetic resonance imaging studies have challenged the accuracy of the clinical diagnosis of pneumonia, and epidemiologic studies are changing our perspective of what causes community acquired pneumonia, especially the role of viral pathogens and the frequent finding of multiple pathogens. The past decade has seen increasing overuse of empiric coverage of meticillin resistant Staphylococcus aureus and antibiotic resistant Gram negative pathogens owing to inappropriate application of guidelines for healthcare associated pneumonia. Optimal treatment remains a matter for debate, especially in very sick patients, including the role of combination antibiotic therapy and corticosteroids. Pneumonia care bundles are being defined to improve outcomes. Increased recognition of both acute and long term cardiac complications is shifting our concept of pneumonia from an acute lung disease to a multisystem problem with adverse chronic health consequences.
The past decade has seen an explosion in the amount of digital information stored in electronic health records (EHR). While primarily designed for archiving patient clinical information and administrative healthcare tasks, many researchers have found secondary use of these records for various clinical informatics tasks. Over the same period, the machine learning community has seen widespread advances in deep learning techniques, which also have been successfully applied to the vast amount of EHR data. In this paper, we review these deep EHR systems, examining architectures, technical aspects, and clinical applications. We also identify shortcomings of current techniques and discuss avenues of future research for EHR-based deep learning.
Purpose To evaluate the efficacy of deep convolutional neural networks (DCNNs) for detecting tuberculosis (TB) on chest radiographs. Materials and Methods Four deidentified HIPAA-compliant datasets were used in this study that were exempted from review by the institutional review board, which consisted of 1007 posteroanterior chest radiographs. The datasets were split into training (68.0%), validation (17.1%), and test (14.9%). Two different DCNNs, AlexNet and GoogLeNet, were used to classify the images as having manifestations of pulmonary TB or as healthy. Both untrained and pretrained networks on ImageNet were used, and augmentation with multiple preprocessing techniques. Ensembles were performed on the best-performing algorithms. For cases where the classifiers were in disagreement, an independent board-certified cardiothoracic radiologist blindly interpreted the images to evaluate a potential radiologist-augmented workflow. Receiver operating characteristic curves and areas under the curve (AUCs) were used to assess model performance by using the DeLong method for statistical comparison of receiver operating characteristic curves. Results The best-performing classifier had an AUC of 0.99, which was an ensemble of the AlexNet and GoogLeNet DCNNs. The AUCs of the pretrained models were greater than that of the untrained models (P < .001). Augmenting the dataset further increased accuracy (P values for AlexNet and GoogLeNet were .03 and .02, respectively). The DCNNs had disagreement in 13 of the 150 test cases, which were blindly reviewed by a cardiothoracic radiologist, who correctly interpreted all 13 cases (100%). This radiologist-augmented approach resulted in a sensitivity of 97.3% and specificity 100%. Conclusion Deep learning with DCNNs can accurately classify TB at chest radiography with an AUC of 0.99. A radiologist-augmented approach for cases where there was disagreement among the classifiers further improved accuracy. (©) RSNA, 2017.
The ability to automatically learn task specific feature representations has led to a huge success of deep learning methods. When large training data is scarce, such as in medical imaging problems, transfer learning has been very effective. In this paper, we systematically investigate the process of transferring a Convolutional Neural Network, trained on ImageNet images to perform image classification, to kidney detection problem in ultrasound images. We study how the detection performance depends on the extent of transfer. We show that a transferred and tuned CNN can outperform a state-of-the-art feature engineered pipeline and a hybridization of these two techniques achieves 20 % higher performance. We also investigate how the evolution of intermediate response images from our network. Finally, we compare these responses to state-of-the-art image processing filters in order to gain greater insight into how transfer learning is able to effectively manage widely varying imaging regimes.