Anamnesio_bot: A Chatbot Prototype Based On Generative Artificial Intelligence For Clinical Anamnesis Study and Simulation

Clinical anamnesis is a fundamental skill in medical education, allowing students to develop competencies in clinical information gathering, patient communication and diagnostic reasoning. Traditionally, teaching anamnesis has been based on interaction with real patients or simulated actors, which poses logistical challenges and access limitations. Generative artificial intelligence offers an innovative alternative for clinical interview simulation through autonomous training in an interactive digital environment. This study describes the design and development of Anamnesio_bot, a prototype chatbot based on Generative Artificial Intelligence designed for clinical anamnesis simulation in medical students. Its implementation was based on the integration of multiple sources of information, including a database of 2000 virtual patients generated with advanced language models, specific medical protocols, standardized clinical history structures and medical teaching principles. The chatbot was programmed to respond in a structured, realistic and concise manner to open-ended questions posed by students, ensuring consistency in the simulation of clinical cases. The development process focused on three key aspects: the generation of a large and diverse database of clinical cases; the optimization of the response algorithms by using predefined medical structures; and the implementation of automated feedback based on anamnesis evaluation criteria. This prototype will undergo evaluation in a real-world setting with students in the near future. However, its design suggests that it may be a useful tool for anamnesis practice in an accessible and flexible environment.