Interaction Between Physicians and Artificial Intelligence Supervisor: Dimitris Kalles (kalles@eap.gr) Objective: This thesis focuses on the development of interaction technologies between physicians and artificial intelligence (AI). Scope: We have developed and are using an online application where users (physicians) interact with AI by asking questions related to their specialty and evaluating the responses based on criteria such as accuracy of information, relevance, clarity, response time, and overall satisfaction with the interaction. The AI responses are generated by a Generative AI system (primarily, ChatGPT). The goal of the thesis is to identify and enhance areas where AI needs to improve its effectiveness, so that physicians can confidently use it as a supportive tool in their daily work. Key Areas to Study: More Accurate and Higher-Quality Responses: Enhance AI response quality by integrating more advanced learning algorithms or connecting to specialized medical databases. Dialogical Interaction with AI: Enable dialogical interaction with AI or AI-assisted interaction with other users (peers). This requires incorporating natural language processing (NLP) algorithms to allow AI to understand and respond to questions naturally. Additionally, include features for discussion, collaboration, and knowledge exchange among users on the platform. AI can suggest medical resources or help verify medical information during discussions. Educational Tool: Use the application as an educational tool by integrating simulation scenarios of complex clinical cases or developing appropriately designed tests to assess users based on material provided by the AI.
References: Key publications focusing on the interaction between physicians and AI include: - Gosbee, J., & Ritchie, E. (1997). Human-computer interaction and medical software development. Interactions, 4(4), 13-18.
- Cai, C. J., Winter, S., Steiner, D., Wilcox, L., & Terry, M. (2019). "Hello AI": uncovering the onboarding needs of medical practitioners for human-AI collaborative decision-making. Proceedings of the ACM on Human-computer Interaction, 3(CSCW), 1-24.
- Karinshak, E., Liu, S. X., Park, J. S., & Hancock, J. T. (2023). Working With AI to Persuade: Examining a Large Language Model's Ability to Generate Pro-Vaccination Messages. Proceedings of the ACM on Human-Computer Interaction, 7(CSCW1), 1-29.
- Nielsen, J. P., von Buchwald, C., & Grønhøj, C. (2023). Validity of the large language model ChatGPT (GPT4) as a patient information source in otolaryngology by a variety of doctors in a tertiary otorhinolaryngology department. Acta Oto-Laryngologica, 1-4.
- Kaarre, J., Feldt, R., Keeling, L. E., Dadoo, S., Zsidai, B., Hughes, J. D., ... & Musahl, V. (2023). Exploring the potential of ChatGPT as a supplementary tool for providing orthopaedic information. Knee Surgery, Sports Traumatology, Arthroscopy, 31(11), 5190-5198.
The current implementation of the application and key information about it can be found at the following website. There, you can use the application and evaluate the results returned by the Artificial Intelligence from your perspective: Dimokratia AI Search Terms: Artificial Intelligence, Doctor-AI Interaction, medical education, user satisfaction |
Σχόλια