AI-Driven Communication Training for Medical Students to Practice Discussing Abnormal Mammogram Results with Patients: Development of a GPT-4 Powered Virtual Simulated Patient (Preprint) DOI Creative Commons

Dan Weisman,

Alanna Sugarman, Yue Huang

и другие.

JMIR Formative Research, Год журнала: 2024, Номер unknown

Опубликована: Авг. 27, 2024

Standardized patients (SPs) prepare medical students for difficult conversations with patients. Despite their value, SP-based simulation training is constrained by available resources and competing clinical demands. Researchers are turning to artificial intelligence large language models, such as generative pretrained transformers, create communication that incorporates virtual simulated (VSPs). GPT-4 a model advance allowing developers design scenarios using text-based prompts instead of relying on branching path simulations prescripted dialogue. These nascent developmental practices have not taken root in the literature guide other researchers developing own simulations. This study aims describe our process lessons learned creating GPT-4-driven VSP. We designed VSP help student learners rehearse discussing abnormal mammography results patient primary care physician (PCP). aimed assess GPT-4's ability generate appropriate responses during spoken provide feedback learner performance. A research team comprised physicians, student, an educator, SP program director, learning experience designer, health researcher conducted study. formative phase in-depth knowledge user interviews informed development, followed development module. The 5 students, PCPs, breast cancer survivors. They then developed authoring software provided GPT-4-enabled initial prompt consisting scenario description, emotional state, expectations It was iteratively refined through agile involving repeated cycles testing, documenting issues, revising prompt. As exploratory feature, used written about performance communicating adherence guidelines conversations. In-depth helped establish timing, mode communication, protocol between PCPs screening process. telephone call discuss diagnostic mammogram indicated need biopsy. Preliminary testing promising. asked sensible questions responded inquiries voice replete inflections. generated successfully identified strengths areas improvement relevant quotes from learner-VSP conversation, but it occasionally misidentified protocols. streamlined facilitated more dynamic, humanlike interactions compared For next steps, we will pilot-test evaluate its feasibility acceptability.

Язык: Английский

Application of Chatbots to Help Patients Self-Manage Diabetes: Systematic Review and Meta-Analysis DOI Creative Commons
Yibo Wu, Jinzi Zhang, Pu Ge

и другие.

Journal of Medical Internet Research, Год журнала: 2024, Номер 26, С. e60380 - e60380

Опубликована: Дек. 3, 2024

Background The number of people with diabetes is on the rise globally. Self-management and health education patients are keys to control diabetes. With development digital therapies artificial intelligence, chatbots have potential provide health-related information improve accessibility effectiveness in field patient self-management. Objective This study systematically reviews current research status self-management support chatbots. Methods A systematic review meta-analysis that can help was conducted. PubMed Web Science databases were searched using keywords around diabetes, chatbots, conversational agents, virtual assistants, more. search period from date creation January 1, 2023. Research articles English fit topic selected, did not or available full text excluded. Results In total, 25 studies included review. terms type, all could be classified as design (n=8, 32%), pilot intervention (n=9, 36%). Many adopted a nonrandomized controlled trial (n=6, 24%), there only 1 (4%) randomized trial. strategy, divided into quantitative (n=10, 40%), mixed qualitative (n=1, 4%). evaluation criteria for chatbot technical performance evaluation, user experience evaluation. Most (n=17, 68%) provided management focused diet, exercise, glucose monitoring, medications, complications, few (n=2, 8%) mental health. found effective lowering blood (mean difference 0.30, 95% CI 0.04-0.55; P=.02) had no significant effect reducing weight 1.41, –2.29 5.11; P=.46) compared baseline. Conclusions Chatbots However, evidence level low, higher (such trials) needed strengthen base. More use strategy fully strengths both research. Appropriate innovative theoretical frameworks should used study. addition, researchers focus personalized user-friendly interactive features well improvements design.

Язык: Английский

Процитировано

5
A scoping review of embodied conversational agents in education: trends and innovations from 2014 to 2024 DOI
Yongxiang Zhang,

Weiwei Pan

Interactive Learning Environments, Год журнала: 2025, Номер unknown, С. 1 - 22

Опубликована: Фев. 24, 2025

Язык: Английский

Процитировано

0
Factors Influencing the Intentions of Patients With Inflammatory Rheumatic Diseases to Use a Digital Human for Medication Information: Qualitative Study DOI Creative Commons
Lex L Haegens, Victor J. B. Huiskes, Bart J. F. van den Bemt

и другие.

Journal of Medical Internet Research, Год журнала: 2025, Номер 27, С. e57697 - e57697

Опубликована: Март 13, 2025

Background Introduction: Patients with inflammatory rheumatic diseases (IRDs) frequently experience drug-related problems (DRPs). DRPs can have negative health consequences and should be addressed promptly to prevent complications. A digital human, which is an embodied conversational agent, could provide medication-related information in a time- place-independent manner support patients preventing decreasing DRPs. Objective This study aims identify factors that influence the intention of IRDs use human retrieve information. Methods qualitative 3 in-person focus groups was conducted among adult diagnosed IRD Netherlands. The prototype innovative tool provides spoken answers questions linked topic, such as (instructional) videos, drug leaflets, other relevant sources. Before group, participants completed preparatory exercise at home become familiar human. semistructured interview guide based on Proctor framework for implementation determinants used about acceptability, adoption, appropriateness, costs, feasibility, fidelity, penetration, sustainability Focus were recorded, transcribed, analyzed thematically. Results included 22 patients, median age 68 (IQR 52-75) years, whom 64% (n=22) female. In total, 6 themes describing influencing patients’ identified: (1) degree individual needs are met; (2) confidence one’s ability human; (3) using resembles interacting (4) technical functioning (5) privacy security; (6) expected benefit Conclusions novel influenced by related each patient’s their features benefits These identified considered during further development increase future adoption. Thereafter, effect applying instrument improve self-management regarding researched.

Язык: Английский

Процитировано

0
Embodied conversational agents for shared decision-making: a scoping review protocol DOI Creative Commons
Hongzhan Jiang, Xuejing Li, Meiqi Meng

и другие.

BMJ Open, Год журнала: 2025, Номер 15(3), С. e095360 - e095360

Опубликована: Март 1, 2025

Introduction Embodied conversational agents (ECAs) are computer-based dialogue systems designed to simulate face-to-face interactions by incorporating human-like physical attributes. Their capacity establish and maintain an empathic relationship in patient positions them as innovative tools that facilitate shared decision-making (SDM). To systematically synthesise the existing evidence concerning development application of ECAs promoting SDM, this protocol delineates a scoping review identify present available within domain. Specifically, outlines will concentrate on key features context including their appearance, mechanisms emotional models, framework, well implementation evaluation clinical settings. Methods analysis The framework established Arksey O’Malley be employed guide process. This systematic retrieval seven databases, PubMed, EMBASE, PsycINFO, Web Science, Cumulative Index Nursing Allied Health Literature, Institute Electrical Electronics Engineers (IEEE) Xplore Digital Library Association for Computing Machinery (ACM) Library. search strategy has been developed conducted across each database, from its inception September 2024. Two researchers conduct literature screening data extraction independently. results organised presented through narrative abstracts, tables and/or figures. Ethics dissemination Ethical approval is not necessary review, it uses have previously collected. Furthermore, obtained reported peer-reviewed journal. Trial registration number Open Science Framework Registries ( https://doi.org/10.17605/OSF.IO/BN3CM ).

Язык: Английский

Процитировано

0
How do you evaluate me? A systematic review of conversational agents supporting self-management in persons with a chronic disease (Preprint) DOI Creative Commons
Tessa F Peerbolte,

Rozanne A van Diggelen,

Pieter van den Haak

и другие.

Journal of Medical Internet Research, Год журнала: 2025, Номер unknown

Опубликована: Фев. 7, 2025

Язык: Английский

Процитировано

0
Designing Virtual Coaching Solutions DOI Creative Commons
Hannes Schlieter, Kai Gand, Thure Weimann

и другие.

Business & Information Systems Engineering, Год журнала: 2024, Номер 66(3), С. 377 - 400

Опубликована: Май 29, 2024

Abstract Especially older persons are prone to disabilities and chronic diseases. These conditions pose a worldwide challenge, leading deteriorating health, economic strain, loss of life, decline in the quality life (QoL). Therefore, healthcare institutions seek enhance their strategies for disease prevention management uphold well-being community. This leads need regain independence improve QoL properly rehabilitate patients. Virtual Coaches (VCs) form Embodied Conversational Agents seen as relevant digital intervention support continuity care. The paper at hand reports on Design Science Research project about implementing VC solution patients' home rehabilitation. study underpins four pivotal design principles: Adaptivity, Coaching Strategy, Multi-user Interface, Sustainable Infrastructure. final artifact was tested with 80 patients which were supported continuing inpatient rehabilitation by using VC. evaluation shows both positive results usability acceptance different use cases impact QoL. Given comprehensive clinical evaluation, system represents safe appealing ensuring medical care access personalized cognitive motor function treatments.

Язык: Английский

Процитировано

2
Mapping the Evolution of Digital Health Research: Bibliometric Overview of Research Hotspots, Trends, and Collaboration of Publications in JMIR (1999-2024) DOI Creative Commons
Jing Hu, Chong Li,

Yunyi Ge

и другие.

Journal of Medical Internet Research, Год журнала: 2024, Номер 26, С. e58987 - e58987

Опубликована: Окт. 17, 2024

While bibliometric studies of individual journals have been conducted, to the best our knowledge, mapping has not yet utilized analyze literature published by Journal Medical Internet Research (JMIR).

Язык: Английский

Процитировано

2
The recent history and near future of digital health in the field of behavioral medicine: an update on progress from 2019 to 2024 DOI Creative Commons
Danielle Arigo, Danielle E. Jake‐Schoffman, Sherry Pagoto

и другие.

Journal of Behavioral Medicine, Год журнала: 2024, Номер unknown

Опубликована: Окт. 28, 2024

Язык: Английский

Процитировано

2
Acceptance, Satisfaction, and Preference with Telemedicine during the COVID-19 Pandemic 2021-2022: Survey among Patients with Chronic Pain (Preprint) DOI Creative Commons
Michael Alexander Harnik, Alina Scheidegger, Larissa Blättler

и другие.

JMIR Formative Research, Год журнала: 2024, Номер 8, С. e53154 - e53154

Опубликована: Март 16, 2024

The COVID-19 pandemic has forced many health care providers to make changes in their treatment, with telemedicine being expanded on a large scale. An earlier study investigated the acceptance of telephone calls but did not record satisfaction treatment or patients' preferences. This warranted follow-up investigate acceptance, satisfaction, and preferences regarding telemedicine, comprising phone consultations, among recipients.

Язык: Английский

Процитировано

1
Educating Healthcare Professionals on AI in Alzheimer's Disease DOI
Shweta Sharma,

Shavez Mansoori,

Manu Goyal

и другие.

Advances in medical technologies and clinical practice book series, Год журнала: 2024, Номер unknown, С. 372 - 388

Опубликована: Июнь 28, 2024

Alzheimer's disease (AD) is the main cause of dementia in older persons. Preclinical AD can persist anywhere from six to ten years, depending on when it first appears. Artificial intelligence (AI) create machines that are capable doing tasks would typically require human intelligence. Machine learning (ML), deep (DL), and natural language processing (NLP) just a few approaches used AI. tools increasingly being predict beginning or course diseases, addition aiding diagnosis. AI has shown be useful tool for improving customized care by providing best possible treatment plans. diagnose provide programs, assist medical practitioners making decisions. But order apply healthcare ethically successfully, problems with bias, security data, need understanding must addressed.

Язык: Английский

Процитировано

1