The project is financed by the The National Center for Research and Development
Grant number: POIR.01.01.01-00-0526/19
Total value: 6.452.485 PLN
Co-financing: 4.859.328 PLN
The aim of the project is to carry out research, development and pre-implementation work, which will result in the introduction of a virtual assistant (chat-bot, assistant) using artificial intelligence to support the process of didactic diagnostics and disease therapy. The solution serves to improve the didactic process of acquiring competences by medical students and young doctors who continue their education. Its task is to improve the quality, attractiveness and cost-effectiveness of the teaching process.
Based on a knowledge base prepared by experts describing selected medical cases and the functionality of interactive student communication with the tool, the assistant will enable learning to make a diagnosis and treatment planning using diagnostic tests: laboratory, image, physical and specialist consultations.
Medical history will reflect the character of spontaneous conversation as much as possible. The student will manage the conversation. He will ask the patient questions and commission appropriate tests to make a diagnosis.
Features and functionality of the designed chat-bot:
- interactive communication using natural language (Polish language);
- interactive creation of “patients” based on rules stored in the knowledge base and models drawing results of diagnostic tests;
- a wizard enabling the creation or modification of clinical cases;
- tracking the student’s progress and the history of his / her interaction with the program, including verifying whether the student asked the required questions and commissioned the required tests, assessment of the time devoted to the various stages of the study;
- assessment of soft competences of the student through the analysis of the form of questions asked;
- assessment of the correctness of the diagnostic path selection made on the basis of analysis of the student’s interaction with the simulated patient.