Towards a Chatbot for Medical Diagnosis Based on Patient Symptoms.

Journal: Studies in health technology and informatics
Published Date:

Abstract

The advent of artificial intelligence has positively transformed many areas of our lives, including the medical field. In this article, we propose the development of a medical diagnosis chatbot based on patients' symptoms, using artificial intelligence as an innovative solution. The aim of this tool is to provide doctors with a preliminary diagnosis based on the symptoms presented by patients. Our proposal involves first building models to be used for predicting diseases, and then using a llama-inspired model to generate the medical report and prescription. The chatbot system was implemented using Python, HTML, CSS and JavaScript, with FastAPI for the RESTful API and PostgreSQL for data management. The results highlight the potential of AI to improve healthcare services, especially in areas where resources are limited.

Authors

  • Kisito Kabore
    Joseph KI ZERBO University, Burkina Faso.
  • Omar Sawadogo
    Joseph KI ZERBO University, Burkina Faso.
  • Yaya Traore
    Joseph KI ZERBO University, Burkina Faso.
  • Julie Thiombiano
    Nazi BONI University, Burkina Faso.