Year: 2025
Type: article
Abstract: The primary tool used in medical diagnosis is anamnesis, which involves the process in which the patient recalls the signs and symptoms related to their pathology. However, conducting anamnesis with other individuals requires extensive training and planning. As a result, the use of virtual patients for anamnesis simulations has been widely adopted to complement practice with real patients. Virtual patients have been broadly accepted as a way to develop students' clinical skills, often surpassing traditional teaching methods. With the development of generative artificial intelligence (AI) in recent years, virtual patients can become more realistic and create more immersive environments for learning anamnesis. However, current virtual patients focus only on predetermined areas or do not yet utilize the latest AI models. In light of this, the present project aims to develop a virtual environment containing a fully functional virtual patient, capable of engaging in natural conversations with an end user. The Unity platform was used to develop the virtual environment, the Google Gemini API to simulate the final patient, and a CSV file containing various pathologies along with their respective symptoms and medical histories. The resulting project successfully achieves its objectives, creating an environment with a virtual patient connected to generative AI that faithfully simulates a patient suffering from a specific pathology, thereby aiding the training of future medical and nursing students.
Institution: Universidade Estadual Paulista (Unesp)
Language: Portuguese
Subfield: Health Informatics
Field: Medicine
Domain: Health Sciences
Open Access status: diamond
OpenAlex ID: W7216795236
