@Article{info:doi/10.2196/59010, author="Tahtali, A. Mohammed and Snijders, P. Chris C. and Dirne, M. Corn{\'e} W. G. and Le Blanc, M. Pascale", title="Prioritizing Trust in Podiatrists' Preference for AI in Supportive Roles Over Diagnostic Roles in Health Care: Qualitative Interview and Focus Group Study", journal="JMIR Hum Factors", year="2025", month="Feb", day="21", volume="12", pages="e59010", keywords="AI's role in health care", keywords="decision-making", keywords="diabetes and podiatrists", keywords="trust", keywords="AI", keywords="artificial intelligence", keywords="qualitative", keywords="foot", keywords="podiatry", keywords="professional", keywords="experience", keywords="attitude", keywords="opinion", keywords="perception", keywords="acceptance", keywords="adoption", keywords="thematic", keywords="focus group", abstract="Background: As artificial intelligence (AI) evolves, its roles have expanded from helping out with routine tasks to making complex decisions, once the exclusive domain of human experts. This shift is pronounced in health care, where AI aids in tasks ranging from image recognition in radiology to personalized treatment plans, demonstrating the potential to, at times, surpass human accuracy and efficiency. Despite AI's accuracy in some critical tasks, the adoption of AI in health care is a challenge, in part because of skepticism about being able to rely on AI decisions. Objective: This study aimed to identify and delve into more effective and acceptable ways of integrating AI into a broader spectrum of health care tasks. Methods: We included 2 qualitative phases to explore podiatrists' views on AI in health care. Initially, we interviewed 9 podiatrists (7 women and 2 men) with a mean age of 41 (SD 12) years and aimed to capture their sentiments regarding the use and role of AI in their work. Subsequently, a focus group with 5 podiatrists (4 women and 1 man) with a mean age of 54 (SD 10) years delved into AI's supportive and diagnostic roles on the basis of the interviews. All interviews were recorded, transcribed verbatim, and analyzed using Atlas.ti and QDA-Miner, using both thematic analysis for broad patterns and framework analysis for structured insights per established guidelines. Results: Our research unveiled 9 themes and 3 subthemes, clarifying podiatrists' nuanced views on AI in health care. Key overlapping insights in the 2 phases included a preference for using AI in supportive roles, such as triage, because of its efficiency and process optimization capabilities. There is a discernible hesitancy toward leveraging AI for diagnostic purposes, driven by concerns regarding its accuracy and the essential nature of human expertise. The need for transparency and explainability in AI systems emerged as a critical factor for fostering trust in both phases. Conclusions: The findings highlight a complex view from podiatrists on AI, showing openness to its application in supportive roles while exercising caution with diagnostic use. This result is consistent with a careful introduction of AI into health care in roles, such as triage, in which there is initial trust, as opposed to roles that ask the AI for a complete diagnosis. Such strategic adoption can mitigate initial resistance, gradually building the confidence to explore AI's capabilities in more nuanced tasks, including diagnostics, where skepticism is currently more pronounced. Adopting AI stepwise could thus enhance trust and acceptance across a broader range of health care tasks, aligning technology integration with professional comfort and patient care standards. ", doi="10.2196/59010", url="https://humanfactors.jmir.org/2025/1/e59010" } @Article{info:doi/10.2196/48633, author="Hassan, Masooma and Kushniruk, Andre and Borycki, Elizabeth", title="Barriers to and Facilitators of Artificial Intelligence Adoption in Health Care: Scoping Review", journal="JMIR Hum Factors", year="2024", month="Aug", day="29", volume="11", pages="e48633", keywords="artificial intelligence", keywords="governance", keywords="health information systems", keywords="artificial intelligence adoption", keywords="system implementation", keywords="health care organizations", keywords="health services", keywords="mobile phone", abstract="Background: Artificial intelligence (AI) use cases in health care are on the rise, with the potential to improve operational efficiency and care outcomes. However, the translation of AI into practical, everyday use has been limited, as its effectiveness relies on successful implementation and adoption by clinicians, patients, and other health care stakeholders. Objective: As adoption is a key factor in the successful proliferation of an innovation, this scoping review aimed at presenting an overview of the barriers to and facilitators of AI adoption in health care. Methods: A scoping review was conducted using the guidance provided by the Joanna Briggs Institute and the framework proposed by Arksey and O'Malley. MEDLINE, IEEE Xplore, and ScienceDirect databases were searched to identify publications in English that reported on the barriers to or facilitators of AI adoption in health care. This review focused on articles published between January 2011 and December 2023. The review did not have any limitations regarding the health care setting (hospital or community) or the population (patients, clinicians, physicians, or health care administrators). A thematic analysis was conducted on the selected articles to map factors associated with the barriers to and facilitators of AI adoption in health care. Results: A total of 2514 articles were identified in the initial search. After title and abstract reviews, 50 (1.99\%) articles were included in the final analysis. These articles were reviewed for the barriers to and facilitators of AI adoption in health care. Most articles were empirical studies, literature reviews, reports, and thought articles. Approximately 18 categories of barriers and facilitators were identified. These were organized sequentially to provide considerations for AI development, implementation, and the overall structure needed to facilitate adoption. Conclusions: The literature review revealed that trust is a significant catalyst of adoption, and it was found to be impacted by several barriers identified in this review. A governance structure can be a key facilitator, among others, in ensuring all the elements identified as barriers are addressed appropriately. The findings demonstrate that the implementation of AI in health care is still, in many ways, dependent on the establishment of regulatory and legal frameworks. Further research into a combination of governance and implementation frameworks, models, or theories to enhance trust that would specifically enable adoption is needed to provide the necessary guidance to those translating AI research into practice. Future research could also be expanded to include attempts at understanding patients' perspectives on complex, high-risk AI use cases and how the use of AI applications affects clinical practice and patient care, including sociotechnical considerations, as more algorithms are implemented in actual clinical environments. ", doi="10.2196/48633", url="https://humanfactors.jmir.org/2024/1/e48633" } @Article{info:doi/10.2196/51972, author="Zainal, Humairah and Hui, Xiao Xin and Thumboo, Julian and Fong, Warren and Yong, Kok Fong", title="Patients' Expectations of Doctors' Clinical Competencies in the Digital Health Care Era: Qualitative Semistructured Interview Study Among Patients", journal="JMIR Hum Factors", year="2024", month="Aug", day="27", volume="11", pages="e51972", keywords="digital health", keywords="clinical competence", keywords="patient engagement", keywords="qualitative research", keywords="Singapore", keywords="mobile phone", abstract="Background: Digital technologies have impacted health care delivery globally, and are increasingly being deployed in clinical practice. However, there is limited research on patients' expectations of doctors' clinical competencies when using digital health care technologies (DHTs) in medical care. Understanding these expectations can reveal competency gaps, enhance patient confidence, and contribute to digital innovation initiatives. Objective: This study explores patients' perceptions of doctors' use of DHTs in clinical care. Using Singapore as a case study, it examines patients' expectations regarding doctors' communication, diagnosis, and treatment skills when using telemedicine, health apps, wearable devices, electronic health records, and artificial intelligence. Methods: Findings were drawn from individual semistructured interviews with patients from outpatient clinics. Participants were recruited using purposive sampling. Data were analyzed qualitatively using thematic analysis. Results: Twenty-five participants from different backgrounds and with various chronic conditions participated in the study. They expected doctors to be adept in handling medical data from apps and wearable devices. For telemedicine, participants expected a level of assessment of their medical conditions akin to in-person consultations. In addition, they valued doctors recognizing when a physical examination was necessary. Interestingly, eye contact was appreciated but deemed nonessential by participants across all age bands when electronic health records were used, as they valued the doctor's efficiency more than eye contact. Nonetheless, participants emphasized the need for empathy throughout the clinical encounter regardless of DHT use. Furthermore, younger participants had a greater expectation for DHT use among doctors compared to older ones, who preferred DHTs as a complement rather than a replacement for clinical skills. The former expected doctors to be knowledgeable about the algorithms, principles, and purposes of DHTs such as artificial intelligence technologies to better assist them in diagnosis and treatment. Conclusions: By identifying patients' expectations of doctors amid increasing health care digitalization, this study highlights that while basic clinical skills remain crucial in the digital age, the role of clinicians needs to evolve with the introduction of DHTs. It has also provided insights into how DHTs can be integrated effectively into clinical settings, aligning with patients' expectations and preferences. Overall, the findings offer a framework for high-income countries to harness DHTs in enhancing health care delivery in the digital era. ", doi="10.2196/51972", url="https://humanfactors.jmir.org/2024/1/e51972" } @Article{info:doi/10.2196/55961, author="Lukkien, M. Dirk R. and Ipakchian Askari, Sima and Stolwijk, E. Nathalie and Hofstede, M. Bob and Nap, Herman Henk and Boon, C. Wouter P. and Peine, Alexander and Moors, M. Ellen H. and Minkman, N. Mirella M.", title="Making Co-Design More Responsible: Case Study on the Development of an AI-Based Decision Support System in Dementia Care", journal="JMIR Hum Factors", year="2024", month="Jul", day="31", volume="11", pages="e55961", keywords="responsible innovation", keywords="co-design", keywords="ethics", keywords="decision support systems", keywords="gerontechnology", keywords="dementia", keywords="long-term care", abstract="Background: Emerging technologies such as artificial intelligence (AI) require an early-stage assessment of potential societal and ethical implications to increase their acceptability, desirability, and sustainability. This paper explores and compares 2 of these assessment approaches: the responsible innovation (RI) framework originating from technology studies and the co-design approach originating from design studies. While the RI framework has been introduced to guide early-stage technology assessment through anticipation, inclusion, reflexivity, and responsiveness, co-design is a commonly accepted approach in the development of technologies to support the care for older adults with frailty. However, there is limited understanding about how co-design contributes to the anticipation of implications. Objective: This paper empirically explores how the co-design process of an AI-based decision support system (DSS) for dementia caregivers is complemented by explicit anticipation of implications. Methods: This case study investigated an international collaborative project that focused on the co-design, development, testing, and commercialization of a DSS that is intended to provide actionable information to formal caregivers of people with dementia. In parallel to the co-design process, an RI exploration took place, which involved examining project members' viewpoints on both positive and negative implications of using the DSS, along with strategies to address these implications. Results from the co-design process and RI exploration were analyzed and compared. In addition, retrospective interviews were held with project members to reflect on the co-design process and RI exploration. Results: Our results indicate that, when involved in exploring requirements for the DSS, co-design participants naturally raised various implications and conditions for responsible design and deployment: protecting privacy, preventing cognitive overload, providing transparency, empowering caregivers to be in control, safeguarding accuracy, and training users. However, when comparing the co-design results with insights from the RI exploration, we found limitations to the co-design results, for instance, regarding the specification, interrelatedness, and context dependency of implications and strategies to address implications. Conclusions: This case study shows that a co-design process that focuses on opportunities for innovation rather than balancing attention for both positive and negative implications may result in knowledge gaps related to social and ethical implications and how they can be addressed. In the pursuit of responsible outcomes, co-design facilitators could broaden their scope and reconsider the specific implementation of the process-oriented RI principles of anticipation and inclusion. ", doi="10.2196/55961", url="https://humanfactors.jmir.org/2024/1/e55961" } @Article{info:doi/10.2196/55716, author="Marcuzzi, Anna and Klevanger, Elisabeth Nina and Aasdahl, Lene and Gismervik, Sigmund and Bach, Kerstin and Mork, Jarle Paul and Nordstoga, Lovise Anne", title="An Artificial Intelligence--Based App for Self-Management of Low Back and Neck Pain in Specialist Care: Process Evaluation From a Randomized Clinical Trial", journal="JMIR Hum Factors", year="2024", month="Jul", day="9", volume="11", pages="e55716", keywords="low back pain", keywords="neck pain", keywords="self-management", keywords="smartphone app", keywords="process evaluation", keywords="focus group", keywords="focus groups", keywords="musculoskeletal", keywords="mHealth", keywords="mobile health", keywords="app", keywords="apps", keywords="applications", keywords="usage", keywords="interview", keywords="interviews", keywords="qualitative", keywords="engagement", abstract="Background: Self-management is endorsed in clinical practice guidelines for the care of musculoskeletal pain. In a randomized clinical trial, we tested the effectiveness of an artificial intelligence--based self-management app (selfBACK) as an adjunct to usual care for patients with low back and neck pain referred to specialist care. Objective: This study is a process evaluation aiming to explore patients' engagement and experiences with the selfBACK app and specialist health care practitioners' views on adopting digital self-management tools in their clinical practice. Methods: App usage analytics in the first 12 weeks were used to explore patients' engagement with the SELFBACK app. Among the 99 patients allocated to the SELFBACK interventions, a purposive sample of 11 patients (aged 27-75 years, 8 female) was selected for semistructured individual interviews based on app usage. Two focus group interviews were conducted with specialist health care practitioners (n=9). Interviews were analyzed using thematic analysis. Results: Nearly one-third of patients never accessed the app, and one-third were low users. Three themes were identified from interviews with patients and health care practitioners: (1) overall impression of the app, where patients discussed the interface and content of the app, reported on usability issues, and described their app usage; (2) perceived value of the app, where patients and health care practitioners described the primary value of the app and its potential to supplement usual care; and (3) suggestions for future use, where patients and health care practitioners addressed aspects they believed would determine acceptance. Conclusions: Although the app's uptake was relatively low, both patients and health care practitioners had a positive opinion about adopting an app-based self-management intervention for low back and neck pain as an add-on to usual care. Both described that the app could reassure patients by providing trustworthy information, thus empowering them to take actions on their own. Factors influencing app acceptance and engagement, such as content relevance, tailoring, trust, and usability properties, were identified. Trial Registration: ClinicalTrials.gov NCT04463043; https://clinicaltrials.gov/study/NCT04463043 ", doi="10.2196/55716", url="https://humanfactors.jmir.org/2024/1/e55716", url="http://www.ncbi.nlm.nih.gov/pubmed/38980710" } @Article{info:doi/10.2196/55964, author="Gabarron, Elia and Larbi, Dillys and Rivera-Romero, Octavio and Denecke, Kerstin", title="Human Factors in AI-Driven Digital Solutions for Increasing Physical Activity: Scoping Review", journal="JMIR Hum Factors", year="2024", month="Jul", day="3", volume="11", pages="e55964", keywords="machine learning", keywords="ML", keywords="artificial intelligence", keywords="AI", keywords="algorithm", keywords="algorithms", keywords="predictive model", keywords="predictive models", keywords="predictive analytics", keywords="predictive system", keywords="practical model", keywords="practical models", keywords="deep learning", keywords="human factors", keywords="physical activity", keywords="physical exercise", keywords="healthy living", keywords="active lifestyle", keywords="exercise", keywords="physically active", keywords="digital health", keywords="mHealth", keywords="mobile health", keywords="app", keywords="apps", keywords="application", keywords="applications", keywords="digital technology", keywords="digital intervention", keywords="digital interventions", keywords="smartphone", keywords="smartphones", keywords="PRISMA", abstract="Background: Artificial intelligence (AI) has the potential to enhance physical activity (PA) interventions. However, human factors (HFs) play a pivotal role in the successful integration of AI into mobile health (mHealth) solutions for promoting PA. Understanding and optimizing the interaction between individuals and AI-driven mHealth apps is essential for achieving the desired outcomes. Objective: This study aims to review and describe the current evidence on the HFs in AI-driven digital solutions for increasing PA. Methods: We conducted a scoping review by searching for publications containing terms related to PA, HFs, and AI in the titles and abstracts across 3 databases---PubMed, Embase, and IEEE Xplore---and Google Scholar. Studies were included if they were primary studies describing an AI-based solution aimed at increasing PA, and results from testing the solution were reported. Studies that did not meet these criteria were excluded. Additionally, we searched the references in the included articles for relevant research. The following data were extracted from included studies and incorporated into a qualitative synthesis: bibliographic information, study characteristics, population, intervention, comparison, outcomes, and AI-related information. The certainty of the evidence in the included studies was evaluated using GRADE (Grading of Recommendations Assessment, Development, and Evaluation). Results: A total of 15 studies published between 2015 and 2023 involving 899 participants aged approximately between 19 and 84 years, 60.7\% (546/899) of whom were female participants, were included in this review. The interventions lasted between 2 and 26 weeks in the included studies. Recommender systems were the most commonly used AI technology in digital solutions for PA (10/15 studies), followed by conversational agents (4/15 studies). User acceptability and satisfaction were the HFs most frequently evaluated (5/15 studies each), followed by usability (4/15 studies). Regarding automated data collection for personalization and recommendation, most systems involved fitness trackers (5/15 studies). The certainty of the evidence analysis indicates moderate certainty of the effectiveness of AI-driven digital technologies in increasing PA (eg, number of steps, distance walked, or time spent on PA). Furthermore, AI-driven technology, particularly recommender systems, seems to positively influence changes in PA behavior, although with very low certainty evidence. Conclusions: Current research highlights the potential of AI-driven technologies to enhance PA, though the evidence remains limited. Longer-term studies are necessary to assess the sustained impact of AI-driven technologies on behavior change and habit formation. While AI-driven digital solutions for PA hold significant promise, further exploration into optimizing AI's impact on PA and effectively integrating AI and HFs is crucial for broader benefits. Thus, the implications for innovation management involve conducting long-term studies, prioritizing diversity, ensuring research quality, focusing on user experience, and understanding the evolving role of AI in PA promotion. ", doi="10.2196/55964", url="https://humanfactors.jmir.org/2024/1/e55964" } @Article{info:doi/10.2196/55399, author="Choudhury, Avishek and Shamszare, Hamid", title="The Impact of Performance Expectancy, Workload, Risk, and Satisfaction on Trust in ChatGPT: Cross-Sectional Survey Analysis", journal="JMIR Hum Factors", year="2024", month="May", day="27", volume="11", pages="e55399", keywords="ChatGPT", keywords="chatbots", keywords="health care", keywords="health care decision-making", keywords="health-related decision-making", keywords="health care management", keywords="decision-making", keywords="user perception", keywords="usability", keywords="usable", keywords="usableness", keywords="usefulness", keywords="artificial intelligence", keywords="algorithms", keywords="predictive models", keywords="predictive analytics", keywords="predictive system", keywords="practical models", keywords="deep learning", keywords="cross-sectional survey", abstract="Background: ChatGPT (OpenAI) is a powerful tool for a wide range of tasks, from entertainment and creativity to health care queries. There are potential risks and benefits associated with this technology. In the discourse concerning the deployment of ChatGPT and similar large language models, it is sensible to recommend their use primarily for tasks a human user can execute accurately. As we transition into the subsequent phase of ChatGPT deployment, establishing realistic performance expectations and understanding users' perceptions of risk associated with its use are crucial in determining the successful integration of this artificial intelligence (AI) technology. Objective: The aim of the study is to explore how perceived workload, satisfaction, performance expectancy, and risk-benefit perception influence users' trust in ChatGPT. Methods: A semistructured, web-based survey was conducted with 607 adults in the United States who actively use ChatGPT. The survey questions were adapted from constructs used in various models and theories such as the technology acceptance model, the theory of planned behavior, the unified theory of acceptance and use of technology, and research on trust and security in digital environments. To test our hypotheses and structural model, we used the partial least squares structural equation modeling method, a widely used approach for multivariate analysis. Results: A total of 607 people responded to our survey. A significant portion of the participants held at least a high school diploma (n=204, 33.6\%), and the majority had a bachelor's degree (n=262, 43.1\%). The primary motivations for participants to use ChatGPT were for acquiring information (n=219, 36.1\%), amusement (n=203, 33.4\%), and addressing problems (n=135, 22.2\%). Some participants used it for health-related inquiries (n=44, 7.2\%), while a few others (n=6, 1\%) used it for miscellaneous activities such as brainstorming, grammar verification, and blog content creation. Our model explained 64.6\% of the variance in trust. Our analysis indicated a significant relationship between (1) workload and satisfaction, (2) trust and satisfaction, (3) performance expectations and trust, and (4) risk-benefit perception and trust. Conclusions: The findings underscore the importance of ensuring user-friendly design and functionality in AI-based applications to reduce workload and enhance user satisfaction, thereby increasing user trust. Future research should further explore the relationship between risk-benefit perception and trust in the context of AI chatbots. ", doi="10.2196/55399", url="https://humanfactors.jmir.org/2024/1/e55399", url="http://www.ncbi.nlm.nih.gov/pubmed/38801658" } @Article{info:doi/10.2196/53559, author="Davis, Joshua and Van Bulck, Liesbet and Durieux, N. Brigitte and Lindvall, Charlotta", title="The Temperature Feature of ChatGPT: Modifying Creativity for Clinical Research", journal="JMIR Hum Factors", year="2024", month="Mar", day="8", volume="11", pages="e53559", keywords="artificial intelligence", keywords="ChatGPT", keywords="clinical communication", keywords="creative", keywords="creativity", keywords="customization", keywords="customize", keywords="customized", keywords="generation", keywords="generative", keywords="language model", keywords="language models", keywords="LLM", keywords="LLMs", keywords="natural language processing", keywords="NLP", keywords="random", keywords="randomness", keywords="tailor", keywords="tailored", keywords="temperature", keywords="text", keywords="texts", keywords="textual", doi="10.2196/53559", url="https://humanfactors.jmir.org/2024/1/e53559", url="http://www.ncbi.nlm.nih.gov/pubmed/38457221" } @Article{info:doi/10.2196/52055, author="Cheah, Hui Min and Gan, Nee Yan and Altice, L. Frederick and Wickersham, A. Jeffrey and Shrestha, Roman and Salleh, Mohd Nur Afiqah and Ng, Seong Kee and Azwa, Iskandar and Balakrishnan, Vimala and Kamarulzaman, Adeeba and Ni, Zhao", title="Testing the Feasibility and Acceptability of Using an Artificial Intelligence Chatbot to Promote HIV Testing and Pre-Exposure Prophylaxis in Malaysia: Mixed Methods Study", journal="JMIR Hum Factors", year="2024", month="Jan", day="26", volume="11", pages="e52055", keywords="artificial intelligence", keywords="acceptability", keywords="chatbot", keywords="feasibility", keywords="HIV prevention", keywords="HIV testing", keywords="men who have sex with men", keywords="MSM", keywords="mobile health", keywords="mHealth", keywords="preexposure prophylaxis", keywords="PrEP", keywords="mobile phone", abstract="Background: The HIV epidemic continues to grow fastest among men who have sex with men (MSM) in Malaysia in the presence of stigma and discrimination. Engaging MSM on the internet using chatbots supported through artificial intelligence (AI) can potentially help HIV prevention efforts. We previously identified the benefits, limitations, and preferred features of HIV prevention AI chatbots and developed an AI chatbot prototype that is now tested for feasibility and acceptability. Objective: This study aims to test the feasibility and acceptability of an AI chatbot in promoting the uptake of HIV testing and pre-exposure prophylaxis (PrEP) in MSM. Methods: We conducted beta testing with 14 MSM from February to April 2022 using Zoom (Zoom Video Communications, Inc). Beta testing involved 3 steps: a 45-minute human-chatbot interaction using the think-aloud method, a 35-minute semistructured interview, and a 10-minute web-based survey. The first 2 steps were recorded, transcribed verbatim, and analyzed using the Unified Theory of Acceptance and Use of Technology. Emerging themes from the qualitative data were mapped on the 4 domains of the Unified Theory of Acceptance and Use of Technology: performance expectancy, effort expectancy, facilitating conditions, and social influence. Results: Most participants (13/14, 93\%) perceived the chatbot to be useful because it provided comprehensive information on HIV testing and PrEP (performance expectancy). All participants indicated that the chatbot was easy to use because of its simple, straightforward design and quick, friendly responses (effort expectancy). Moreover, 93\% (13/14) of the participants rated the overall chatbot quality as high, and all participants perceived the chatbot as a helpful tool and would refer it to others. Approximately 79\% (11/14) of the participants agreed they would continue using the chatbot. They suggested adding a local language (ie, Bahasa Malaysia) to customize the chatbot to the Malaysian context (facilitating condition) and suggested that the chatbot should also incorporate more information on mental health, HIV risk assessment, and consequences of HIV. In terms of social influence, all participants perceived the chatbot as helpful in avoiding stigma-inducing interactions and thus could increase the frequency of HIV testing and PrEP uptake among MSM. Conclusions: The current AI chatbot is feasible and acceptable to promote the uptake of HIV testing and PrEP. To ensure the successful implementation and dissemination of AI chatbots in Malaysia, they should be customized to communicate in Bahasa Malaysia and upgraded to provide other HIV-related information to improve usability, such as mental health support, risk assessment for sexually transmitted infections, AIDS treatment, and the consequences of contracting HIV. ", doi="10.2196/52055", url="https://humanfactors.jmir.org/2024/1/e52055", url="http://www.ncbi.nlm.nih.gov/pubmed/38277206" } @Article{info:doi/10.2196/53378, author="Chen, Hongbo and Cohen, Eldan and Wilson, Dulaney and Alfred, Myrtede", title="A Machine Learning Approach with Human-AI Collaboration for Automated Classification of Patient Safety Event Reports: Algorithm Development and Validation Study", journal="JMIR Hum Factors", year="2024", month="Jan", day="25", volume="11", pages="e53378", keywords="accident", keywords="accidents", keywords="black box", keywords="classification", keywords="classifier", keywords="collaboration", keywords="design", keywords="document", keywords="documentation", keywords="documents", keywords="explainability", keywords="explainable", keywords="human-AI collaboration", keywords="human-AI", keywords="human-computer", keywords="human-machine", keywords="incident reporting", keywords="interface design", keywords="interface", keywords="interpretable", keywords="LIME", keywords="machine learning", keywords="patient safety", keywords="predict", keywords="prediction", keywords="predictions", keywords="predictive", keywords="report", keywords="reporting", keywords="safety", keywords="text", keywords="texts", keywords="textual", keywords="artificial intelligence", abstract="Background: Adverse events refer to incidents with potential or actual harm to patients in hospitals. These events are typically documented through patient safety event (PSE) reports, which consist of detailed narratives providing contextual information on the occurrences. Accurate classification of PSE reports is crucial for patient safety monitoring. However, this process faces challenges due to inconsistencies in classifications and the sheer volume of reports. Recent advancements in text representation, particularly contextual text representation derived from transformer-based language models, offer a promising solution for more precise PSE report classification. Integrating the machine learning (ML) classifier necessitates a balance between human expertise and artificial intelligence (AI). Central to this integration is the concept of explainability, which is crucial for building trust and ensuring effective human-AI collaboration. Objective: This study aims to investigate the efficacy of ML classifiers trained using contextual text representation in automatically classifying PSE reports. Furthermore, the study presents an interface that integrates the ML classifier with the explainability technique to facilitate human-AI collaboration for PSE report classification. Methods: This study used a data set of 861 PSE reports from a large academic hospital's maternity units in the Southeastern United States. Various ML classifiers were trained with both static and contextual text representations of PSE reports. The trained ML classifiers were evaluated with multiclass classification metrics and the confusion matrix. The local interpretable model-agnostic explanations (LIME) technique was used to provide the rationale for the ML classifier's predictions. An interface that integrates the ML classifier with the LIME technique was designed for incident reporting systems. Results: The top-performing classifier using contextual representation was able to obtain an accuracy of 75.4\% (95/126) compared to an accuracy of 66.7\% (84/126) by the top-performing classifier trained using static text representation. A PSE reporting interface has been designed to facilitate human-AI collaboration in PSE report classification. In this design, the ML classifier recommends the top 2 most probable event types, along with the explanations for the prediction, enabling PSE reporters and patient safety analysts to choose the most suitable one. The LIME technique showed that the classifier occasionally relies on arbitrary words for classification, emphasizing the necessity of human oversight. Conclusions: This study demonstrates that training ML classifiers with contextual text representations can significantly enhance the accuracy of PSE report classification. The interface designed in this study lays the foundation for human-AI collaboration in the classification of PSE reports. The insights gained from this research enhance the decision-making process in PSE report classification, enabling hospitals to more efficiently identify potential risks and hazards and enabling patient safety analysts to take timely actions to prevent patient harm. ", doi="10.2196/53378", url="https://humanfactors.jmir.org/2024/1/e53378", url="http://www.ncbi.nlm.nih.gov/pubmed/38271086" } @Article{info:doi/10.2196/47031, author="Shevtsova, Daria and Ahmed, Anam and Boot, A. Iris W. and Sanges, Carmen and Hudecek, Michael and Jacobs, L. John J. and Hort, Simon and Vrijhoef, M. Hubertus J.", title="Trust in and Acceptance of Artificial Intelligence Applications in Medicine: Mixed Methods Study", journal="JMIR Hum Factors", year="2024", month="Jan", day="17", volume="11", pages="e47031", keywords="trust", keywords="acceptance", keywords="artificial intelligence", keywords="medicine", keywords="mixed methods", keywords="rapid review", keywords="survey", abstract="Background: Artificial intelligence (AI)--powered technologies are being increasingly used in almost all fields, including medicine. However, to successfully implement medical AI applications, ensuring trust and acceptance toward such technologies is crucial for their successful spread and timely adoption worldwide. Although AI applications in medicine provide advantages to the current health care system, there are also various associated challenges regarding, for instance, data privacy, accountability, and equity and fairness, which could hinder medical AI application implementation. Objective: The aim of this study was to identify factors related to trust in and acceptance of novel AI-powered medical technologies and to assess the relevance of those factors among relevant stakeholders. Methods: This study used a mixed methods design. First, a rapid review of the existing literature was conducted, aiming to identify various factors related to trust in and acceptance of novel AI applications in medicine. Next, an electronic survey including the rapid review--derived factors was disseminated among key stakeholder groups. Participants (N=22) were asked to assess on a 5-point Likert scale (1=irrelevant to 5=relevant) to what extent they thought the various factors (N=19) were relevant to trust in and acceptance of novel AI applications in medicine. Results: The rapid review (N=32 papers) yielded 110 factors related to trust and 77 factors related to acceptance toward AI technology in medicine. Closely related factors were assigned to 1 of the 19 overarching umbrella factors, which were further grouped into 4 categories: human-related (ie, the type of institution AI professionals originate from), technology-related (ie, the explainability and transparency of AI application processes and outcomes), ethical and legal (ie, data use transparency), and additional factors (ie, AI applications being environment friendly). The categorized 19 umbrella factors were presented as survey statements, which were evaluated by relevant stakeholders. Survey participants (N=22) represented researchers (n=18, 82\%), technology providers (n=5, 23\%), hospital staff (n=3, 14\%), and policy makers (n=3, 14\%). Of the 19 factors, 16 (84\%) human-related, technology-related, ethical and legal, and additional factors were considered to be of high relevance to trust in and acceptance of novel AI applications in medicine. The patient's gender, age, and education level were found to be of low relevance (3/19, 16\%). Conclusions: The results of this study could help the implementers of medical AI applications to understand what drives trust and acceptance toward AI-powered technologies among key stakeholders in medicine. Consequently, this would allow the implementers to identify strategies that facilitate trust in and acceptance of medical AI applications among key stakeholders and potential users. ", doi="10.2196/47031", url="https://humanfactors.jmir.org/2024/1/e47031", url="http://www.ncbi.nlm.nih.gov/pubmed/38231544" } @Article{info:doi/10.2196/49788, author="Wang, Bijun and Asan, Onur and Mansouri, Mo", title="Perspectives of Patients With Chronic Diseases on Future Acceptance of AI--Based Home Care Systems: Cross-Sectional Web-Based Survey Study", journal="JMIR Hum Factors", year="2023", month="Nov", day="6", volume="10", pages="e49788", keywords="consumer informatics", keywords="artificial intelligence", keywords="AI", keywords="technology acceptance model", keywords="adoption", keywords="chronic", keywords="motivation", keywords="cross-sectional", keywords="home care", keywords="perception", keywords="perceptions", keywords="attitude", keywords="attitudes", keywords="intent", keywords="intention", abstract="Background: Artificial intelligence (AI)--based home care systems and devices are being gradually integrated into health care delivery to benefit patients with chronic diseases. However, existing research mainly focuses on the technical and clinical aspects of AI application, with an insufficient investigation of patients' motivation and intention to adopt such systems. Objective: This study aimed to examine the factors that affect the motivation of patients with chronic diseases to adopt AI-based home care systems and provide empirical evidence for the proposed research hypotheses. Methods: We conducted a cross-sectional web-based survey with 222 patients with chronic diseases based on a hypothetical scenario. Results: The results indicated that patients have an overall positive perception of AI-based home care systems. Their attitudes toward the technology, perceived usefulness, and comfortability were found to be significant factors encouraging adoption, with a clear understanding of accountability being a particularly influential factor in shaping patients' attitudes toward their motivation to use these systems. However, privacy concerns persist as an indirect factor, affecting the perceived usefulness and comfortability, hence influencing patients' attitudes. Conclusions: This study is one of the first to examine the motivation of patients with chronic diseases to adopt AI-based home care systems, offering practical insights for policy makers, care or technology providers, and patients. This understanding can facilitate effective policy formulation, product design, and informed patient decision-making, potentially improving the overall health status of patients with chronic diseases. ", doi="10.2196/49788", url="https://humanfactors.jmir.org/2023/1/e49788", url="http://www.ncbi.nlm.nih.gov/pubmed/37930780" } @Article{info:doi/10.2196/48476, author="Vijayakumar, Smrithi and Lee, Vien V. and Leong, Ying Qiao and Hong, Jung Soo and Blasiak, Agata and Ho, Dean", title="Physicians' Perspectives on AI in Clinical Decision Support Systems: Interview Study of the CURATE.AI Personalized Dose Optimization Platform", journal="JMIR Hum Factors", year="2023", month="Oct", day="30", volume="10", pages="e48476", keywords="artificial intelligence", keywords="AI", keywords="clinical decision support system", keywords="CDSS", keywords="adoption", keywords="perception", keywords="decision support", keywords="acceptance", keywords="perspective", keywords="perspectives", keywords="opinion", keywords="attitude", keywords="qualitative", keywords="focus", keywords="interview", keywords="interviews", abstract="Background: Physicians play a key role in integrating new clinical technology into care practices through user feedback and growth propositions to developers of the technology. As physicians are stakeholders involved through the technology iteration process, understanding their roles as users can provide nuanced insights into the workings of these technologies that are being explored. Therefore, understanding physicians' perceptions can be critical toward clinical validation, implementation, and downstream adoption. Given the increasing prevalence of clinical decision support systems (CDSSs), there remains a need to gain an in-depth understanding of physicians' perceptions and expectations toward their downstream implementation. This paper explores physicians' perceptions of integrating CURATE.AI, a novel artificial intelligence (AI)--based and clinical stage personalized dosing CDSSs, into clinical practice. Objective: This study aims to understand physicians' perspectives of integrating CURATE.AI for clinical work and to gather insights on considerations of the implementation of AI-based CDSS tools. Methods: A total of 12 participants completed semistructured interviews examining their knowledge, experience, attitudes, risks, and future course of the personalized combination therapy dosing platform, CURATE.AI. Interviews were audio recorded, transcribed verbatim, and coded manually. The data were thematically analyzed. Results: Overall, 3 broad themes and 9 subthemes were identified through thematic analysis. The themes covered considerations that physicians perceived as significant across various stages of new technology development, including trial, clinical implementation, and mass adoption. Conclusions: The study laid out the various ways physicians interpreted an AI-based personalized dosing CDSS, CURATE.AI, for their clinical practice. The research pointed out that physicians' expectations during the different stages of technology exploration can be nuanced and layered with expectations of implementation that are relevant for technology developers and researchers. ", doi="10.2196/48476", url="https://humanfactors.jmir.org/2023/1/e48476", url="http://www.ncbi.nlm.nih.gov/pubmed/37902825" } @Article{info:doi/10.2196/47564, author="Shahsavar, Yeganeh and Choudhury, Avishek", title="User Intentions to Use ChatGPT for Self-Diagnosis and Health-Related Purposes: Cross-sectional Survey Study", journal="JMIR Hum Factors", year="2023", month="May", day="17", volume="10", pages="e47564", keywords="human factors", keywords="behavioral intention", keywords="chatbots", keywords="health care", keywords="integrated diagnostics", keywords="use", keywords="ChatGPT", keywords="artificial intelligence", keywords="users", keywords="self-diagnosis", keywords="decision-making", keywords="integration", keywords="willingness", keywords="policy", abstract="Background: With the rapid advancement of artificial intelligence (AI) technologies, AI-powered chatbots, such as Chat Generative Pretrained Transformer (ChatGPT), have emerged as potential tools for various applications, including health care. However, ChatGPT is not specifically designed for health care purposes, and its use for self-diagnosis raises concerns regarding its adoption's potential risks and benefits. Users are increasingly inclined to use ChatGPT for self-diagnosis, necessitating a deeper understanding of the factors driving this trend. Objective: This study aims to investigate the factors influencing users' perception of decision-making processes and intentions to use ChatGPT for self-diagnosis and to explore the implications of these findings for the safe and effective integration of AI chatbots in health care. Methods: A cross-sectional survey design was used, and data were collected from 607 participants. The relationships between performance expectancy, risk-reward appraisal, decision-making, and intention to use ChatGPT for self-diagnosis were analyzed using partial least squares structural equation modeling (PLS-SEM). Results: Most respondents were willing to use ChatGPT for self-diagnosis (n=476, 78.4\%). The model demonstrated satisfactory explanatory power, accounting for 52.4\% of the variance in decision-making and 38.1\% in the intent to use ChatGPT for self-diagnosis. The results supported all 3 hypotheses: The higher performance expectancy of ChatGPT ($\beta$=.547, 95\% CI 0.474-0.620) and positive risk-reward appraisals ($\beta$=.245, 95\% CI 0.161-0.325) were positively associated with the improved perception of decision-making outcomes among users, and enhanced perception of decision-making processes involving ChatGPT positively impacted users' intentions to use the technology for self-diagnosis ($\beta$=.565, 95\% CI 0.498-0.628). Conclusions: Our research investigated factors influencing users' intentions to use ChatGPT for self-diagnosis and health-related purposes. Even though the technology is not specifically designed for health care, people are inclined to use ChatGPT in health care contexts. Instead of solely focusing on discouraging its use for health care purposes, we advocate for improving the technology and adapting it for suitable health care applications. Our study highlights the importance of collaboration among AI developers, health care providers, and policy makers in ensuring AI chatbots' safe and responsible use in health care. By understanding users' expectations and decision-making processes, we can develop AI chatbots, such as ChatGPT, that are tailored to human needs, providing reliable and verified health information sources. This approach not only enhances health care accessibility but also improves health literacy and awareness. As the field of AI chatbots in health care continues to evolve, future research should explore the long-term effects of using AI chatbots for self-diagnosis and investigate their potential integration with other digital health interventions to optimize patient care and outcomes. In doing so, we can ensure that AI chatbots, including ChatGPT, are designed and implemented to safeguard users' well-being and support positive health outcomes in health care settings. ", doi="10.2196/47564", url="https://humanfactors.jmir.org/2023/1/e47564", url="http://www.ncbi.nlm.nih.gov/pubmed/37195756" }