A system for automated general medical diagnosis using Bayesian networks

Area of Science:

• Computer-assisted diagnostics
• Medical informatics
• Artificial intelligence in healthcare

Background:

• General medical diagnosis can be complex and time-consuming.
• Accessible self-diagnostic tools can empower individuals to seek timely medical attention.
• Existing diagnostic systems may lack scalability or user-friendliness.

Purpose of the Study:

• To present a computer-assisted diagnostic system for general medical diagnosis.
• To provide a free, online, self-diagnostic tool for the public.
• To encourage users to consult a doctor when necessary.

Main Methods:

• Development of a diagnostic system using Bayesian network methodology.
• Creation of a medical database by expert professionals.
• Implementation of a distributed software architecture addressing scalability.
• Evaluation of computation performance.

Main Results:

• The system processed over 97,000 diagnoses.
• Analysis revealed stress and unhealthy lifestyle as prevalent health concerns.
• Identified common health problems among the young generation in Poland.

Conclusions:

• The developed Bayesian network system offers a scalable and efficient approach to computer-assisted diagnosis.
• The tool effectively highlights lifestyle-related health issues in young populations.
• The system serves as a valuable educational resource, promoting proactive health management.