Figure 3-I - uploaded by Pariya Kashfi
Content may be subject to copyright.
nformation modeling
Source publication
A clinical decision support system (CDSS) is an interactive application that is used to facilitate the process of decisionmaking in a clinical context. Developing a usable CDSS is a challenging process; mostly because of the complex nature of domain knowledge and the context of use of those systems. This paper describes how a user centered design (...
Contexts in source publication
Context 1
... experts were involved in this process to mi- nimize the subjectivity of the design and to be as broad as pos- sible in collecting knowledge. A sample mind map is depicted in Figure 3. ...
Context 2
... changes in the domain concept model should be checked from the GUI point of view. Therefore, as depicted in Figure 3, in each iteration, there should be an input from the left hand side process (domain concept model) to the right hand one (GUI). In other words, after each domain concept modeling iteration, the necessity of a new iteration for GUI design should be checked. ...
Citations
... However, these guidelines were devised according to a review of publications that described barriers to CDS use (Trinkley et al., 2020), and physicians' preferences were not part of the production of these guidelines. In addition, CDS developers have generally solicited healthcare professionals' feedback on clinical content or validation in the form of usability testing after the initial prototyping of the CDS design (Kashfi et al., 2010;Marcy et al., 2008 ), ignoring other aspects of the user experience such as emotionally appealing relationships between the user and the product (Desmet et al., 2007) . In addition, little literature describes physician expectations and preferences before designing these systems as an essential phase in the user-centered designs. ...
Research has focused heavily on the study of Clinical Decision Support Systems. However, CDS systems have generally had little impact on clinical practice. One of the most important reasons is the lack of human-computer interaction (HCI) considerations in designing these systems. Although physicians play an essential role in healthcare decision-making, there is little literature describing physicians' expectations and preferences prior to the development of these systems, which is an essential phase in user-centered design. This study aims to answer the following research question: What do physicians expect of interacting with future clinical decision support systems? An exploratory qualitative study was conducted, and data were collected by interviewing 9 physicians practicing in Sweden. A thematic analysis was used for data analysis, and the findings are four themes: 1) physicians' Expectations related to clinical practice; 2) physicians' expectations related to physician-patient relationship; 3) physicians' expectations related to the physician's role 4) physicians' expectations related to CDS governance. The research findings contribute to the knowledge of Anticipated UX in the context of healthcare and CDS systems. The empirical findings on potential user expectations are valuable for understanding the diversity of user experience and user expectations as phenomena in the specific domain of CDS systems. Service designers can utilize and build on the empirical findings to develop positive user experiences of future CDS systems.
... They stress the importance of a dialogue between researchers and the committee Artificial Intelligence for Engineering Design, Analysis and Manufacturing 3 (Bruckman, 2014;Munteanu et al., 2015;Wilson et al., 2017), as well as a stronger acknowledgement of the flexible nature of participatory design research (Goodyear-Smith et al., 2015;Munteanu et al., 2015). Therefore, ongoing collaboration with clinicians is required in order to understand results and to find ways for these results to be implemented, which can lead to a bottleneck that slows down the process of understanding and applying design knowledge in the context (Kashfi, 2010). It is important to think about the ethical implications of collecting, analyzing, and visualizing personal data and it is of utmost importance to enforce the right to privacy, and the right to the protection of personal data of individuals (Lynskey, 2015). ...
Data-enabled design (DED) is a promising new methodology for designing with users from within their own context in an iterative and hands-on fashion. However, the agile and flexible qualities of the methodology do not directly translate to every context. In this article, we reflect on the design process of an intelligent ecosystem, called ORBIT, and a proposed evaluative study planned with it. This was part of a DED project in collaboration with a medical hospital to study the post-operative behavior in the (remote) context of bariatric patients. The design and preparation of this project and the process towards an eventual study rejection from the medical ethical committee (METC) provide rich insights into (1) what it means to conduct DED research in a clinical context, and (2) where the boundaries of the method might lie in this specific application area. We highlight insights from carefully designing the substantial infrastructure for the study, and how different aspects of DED translated less easily to the clinical context. We analyze the proposed study setup through the lenses of several modifications we made to DED and further reflect on how to expand and scale up the methodology and adapt the process for the clinical context.
... Common methods used in user-centred design include personas, think aloud evaluations and workshop with users [11]. However, previous research has shown that usercentred principles are not systematically integrated and followed by HIT vendors [3,8], even though there have been attempts at achieving such an integration [12]. There are for instance efforts in adapting user-centred design methods to public organizations in general [13], as well as more specifically to health care organizations, where these efforts have for example resulted in use scenarios [14] and multidisciplinary thematic seminars [15]. ...
Often, Health Information Technology (HIT) in hospitals consists of off the shelf systems that are configured and implemented by IT department workers. This means that these employees have a significant impact of the usability of HIT systems. Nonetheless, we currently do not know how IT department workers work. This prevents us from formulating educated recommendations aimed at improving HIT usability, known to be poor, especially from nurses’ perspective. In this paper, we hence present the results from an interview study, shedding light on 1) the communication channels that exist between nurses and IT department at a large public hospital in Sweden, and 2) the problems that undermine system-related communication between these two groups. Our findings stress the need for successful two-way communication between nurses and IT department in order to improve the usability of HIT in use.
... The development of a CDSS is complex, costly and interdisciplinary [6]. For the development of the ENSURE prototype the user-centered design approach was applied [7]. To add value for the end users the Five Rights according to Campbell [8] were included in the planning of the CDSS intervention. ...
The development of clinical decision support systems (CDSS) is complex and requires user-centered planning of assistive interventions. Especially in the setting of emergency care requiring time-critical decisions and interventions, it is important to adapt a CDSS to the needs of the user in terms of acceptance, usability and utility. In the so-called ENSURE project, a user-centered approach was applied to develop the CDSS intervention. In the context of this paper, we present a path to the first mockup development for a CDSS interface by addressing Campbell’s Five Rights within the CDSS workflow.
... To prevent the physical separation of navigation information and the surgical site, we propose the use of a tool-mounted interface (TMI). The concept of the TMI is an outcome of user-centered design [26]. The navigation concept comprises a handle, a shaft for tool guidance, a tracking array and a mount for the visualization unit ( Figure 3). ...
Screw placement in the correct angular trajectory is one of the most intricate tasks during spinal fusion surgery. Due to the crucial role of pedicle screw placement for the outcome of the operation, spinal navigation has been introduced into the clinical routine. Despite its positive effects on the precision and safety of the surgical procedure, local separation of the navigation information and the surgical site, combined with intricate visualizations, limit the benefits of the navigation systems. Instead of a tech-driven design, a focus on usability is required in new research approaches to enable advanced and effective visualizations. This work presents a new tool-mounted interface (TMI) for pedicle screw placement. By fixing a TMI onto the surgical instrument, physical de-coupling of the anatomical target and navigation information is resolved. A total of 18 surgeons participated in a usability study comparing the TMI to the state-of-the-art visualization on an external screen. With the usage of the TMI, significant improvements in system usability (Kruskal–Wallis test p < 0.05) were achieved. A significant reduction in mental demand and overall cognitive load, measured using a NASA-TLX (p < 0.05), were observed. Moreover, a general improvement in performance was shown by means of the surgical task time (one-way ANOVA p < 0.001).
... Der Demonstrator PosiThera wurde nach dem User Centered Design entwickelt [13]. Bereits frühzeitig im Projekt wurden mittels Expertenworkshops wesentliche Anforderungen bidirektional mit Experten aus der Versorgung diskutiert und aufgenommen [4] ...
Hinter dem populärwissenschaftlichen Begriff künstliche Intelligenz (KI) verbirgt sich eine Familie von unterschiedlichen technischen Verfahren, welche Daten – in Vielgestalt etwa aus Registern, Leitlinien, Erfahrungswissen oder Heuristiken – zum Zweck der Wissensgenerierung und -ableitung verarbeiten [1,2]. Im Kontext des Gesundheitswesens ergeben sich viele Szenarien für den Einsatz von KI-Verfahren in Form von klinischen Entscheidungsunterstützungssystemen (CDSS), um Praktiker zu unterstützen. So zeigen sich etwa in der Dermatologie vielversprechende Möglichkeiten mithilfe von KI, Melanom-Bilder effizient zu analysieren und mithilfe der Analyseergebnisse den Praktiker zu unterstützen [3]. Auch in der Pflege sind Szenarien für den Einsatz von KI-Verfahren denkbar und erscheinen lohnenswert. So könnten beispielsweise unerfahrene Mitarbeiter*innen von einem KI- gestützten »OnBoarding« profitieren, welches ähnlich einer Fahrassistenz im Automobil Hinweise und Handlungsempfehlungen liefert. Aber auch die Weiterbildung des Pflegepersonals könnte durch solche Systeme unterstützt werden. Ein CDSS, welches Hinweise zu wissenschaftlichen Erkenntnissen im aktuellen Versorgungskontext liefert, wäre eine Bereicherung, um das Pflegepersonal berufsbegleitend weiterzubilden. Die Vision einer sozio-technischen Partnerschaft zwischen Pflegekraft und KI-System könnte eine Chance bieten, Personal kognitiv zu entlasten und gleichzeitig die Qualität der pflegerischen Prozesse zu verbessern.
Im Rahmen des BMBF-geförderten Projektes PosiThera wurde ein gleichnamiger Demonstrator als ein KI-basiertes CDSS entwickelt, welcher die Behandlung chronischer Wunden unterstützen soll, die durch ihre Interprofessionalität charakterisiert ist. Unterschiedliche Professionen, Disziplinen und Versorgungssektoren arbeiten bei Diagnostik und Therapie zusammen. In diesem Kontext ist es besonders wichtig, gemeinsame Informationsströme zu bedienen, aber auch fachübergreifendes Wissen in die Behandlung einfließen zu lassen. Der Demonstrator PosiThera setzt an diesen Strukturen an und bietet eine KI-basierte Plattform, welche unter Einbeziehung der verteilten Behandlungsdaten Hinweise zu Diagnostik und Therapie einer chronischen Wunde liefert. In der Projektlaufzeit wurden Expertenworkshops ausgerichtet, in denen identifiziert werden konnte, dass der Einsatz von Künstlicher Intelligenz besonders erstrebenswert in den Anwendungsfällen des »diagnostischen Prozesses« und während des »Wundmonitorings« ist [4].
... Acceptance can arise if users are involved as early as possible in the development process of a software. User-centered design is based on this concept [6,7]. ...
The PosiThera project focuses on the management of chronic wounds, which is multi-professional and multi-disciplinary. For this context, a software prototype was developed in the project, which is intended to support medical and nursing staff with the assistance of artificial intelligence. In accordance with the user-centred design, national workshops were held at the beginning of the project with the involvement of domain experts in wound care in order to identify requirements and use cases of IT systems in wound care, with a focus on AI. In this study, the focus was on involving nursing and nursing science staff in testing the software prototype to gain insights into its functionality and usability. The overarching goal of the iterative testing and adaptation process is to further develop the prototype in a way that is close to care.
... Despite this possibilities, as has also been mentioned above, we are going to dedicate a strict user-centred design approach, thus adapting the final PRESTO app and platform to the needs and requirements of end-users. 65 If successfully validated and its effectiveness is confirmed, we have set out in advance a valorization, cooperation and transferability plan to the public health system. ...
Background:
About 30 to 50% of Primary Care (PC) users in Spain suffer mental health problems, mostly mild to moderate anxious and depressive symptoms, which account for 2% of Spain's total Gross domestic product and 50% of the costs associated to all mental disorders. Mobile health tools have demonstrated to cost-effectively reduce anxious and depressive symptoms while machine learning (ML) techniques have shown to accurately detect severe cases. The main aim of this project is to develop a comprehensive ML digital support platform (PRESTO) to cost-effectively screen, assess, triage, and provide personalized treatments for anxious and depressive symptoms in PC.
Methods:
The project will be carried out in 3 complementary phases: First, a ML predictive severity model will be built based on all the cases referred to the PC mental health support programme during the last 5 years in Catalonia. Simultaneously, a smartphone app to monitor and deliver psychological interventions for anxiety and depressive symptoms will be developed and tested in a clinical trial. Finally, the ML models and the app will be integrated in a comprehensive decision-support platform (PRESTO) which will triage and assign to each patient a specific intervention based on individual personal and clinical characteristics. The effectiveness of PRESTO to reduce waiting times in receiving mental healthcare will be tested in a stepped-wedge cluster randomized controlled trial in 5 PC centres.
Discussion:
PRESTO will offer timely and personalized cost-effective mental health treatment to people with mild to moderate anxious and depressive symptoms. This will result in a reduction of the burden of mental health problems in PC and on society as a whole.
Trial registration:
The project and their clinical trials were registered in Clinical Trials.gov: NCT04559360 (September 2020).
... Up to 95% of CDS tools are dismissed [5], 52%-66% improve process outcomes such as appropriate drug selection, and only 21%-43% lead to improvements in clinical outcomes [6][7][8][9]. CDS design best practices may be a way to improve the impact of CDS tools [10,11]. As a framework, CDS design best practices are intended to provide a narrative representation of the key determinants that influence the success of CDS tools [12][13][14][15][16][17][18]. ...
Background
Clinical decision support (CDS) design best practices are intended to provide a narrative representation of factors that influence the success of CDS tools. However, they provide incomplete direction on evidence-based implementation principles.
Objective
This study aims to describe an integrated approach toward applying an existing implementation science (IS) framework with CDS design best practices to improve the effectiveness, sustainability, and reproducibility of CDS implementations.
Methods
We selected the Practical Robust Implementation and Sustainability Model (PRISM) IS framework. We identified areas where PRISM and CDS design best practices complemented each other and defined methods to address each. Lessons learned from applying these methods were then used to further refine the integrated approach.
Results
Our integrated approach to applying PRISM with CDS design best practices consists of 5 key phases that iteratively interact and inform each other: multilevel stakeholder engagement, designing the CDS, design and usability testing, thoughtful deployment, and performance evaluation and maintenance. The approach is led by a dedicated implementation team that includes clinical informatics and analyst builder expertise.
Conclusions
Integrating PRISM with CDS design best practices extends user-centered design and accounts for the multilevel, interacting, and dynamic factors that influence CDS implementation in health care. Integrating PRISM with CDS design best practices synthesizes the many known contextual factors that can influence the success of CDS tools, thereby enhancing the reproducibility and sustainability of CDS implementations. Others can adapt this approach to their situation to maximize and sustain CDS implementation success.
... To avoid physical de-coupling of the navigation and the surgical situs, we propose the use of a hand-held navigation device (HND). This HND is a custom-made device, developed as a result of user-centered design [12]. It consists of a handle, a shaft for tool guidance, a tracking array and a visualization unit holder. ...
During spinal fusion surgery, the orientation of the pedicle screw in the right angle plays a crucial role for the outcome of the operation. Local separation of navigation information and the surgical situs, in combination with intricate visualizations, can limit the benefits of surgical navigation systems. The present study addresses these problems by proposing a hand-held navigation device (HND) for pedicle screw placement. The in-situ visualization of graphically reduced interfaces, and the simple integration of the device into the surgical work flow, allow the surgeon to position the tool while keeping sight of the anatomical target. 18 surgeons participated in a study comparing the HND to the state-of-the-art visualization on an external screen. Our approach revealed significant improvements in mental demand and overall cognitive load, measured using NASA-TLX (p<0.05). Moreover, surgical time (One-Way ANOVA p<0.001) and system usability (Kruskal-Wallis test p<0.05) were significantly improved.
