Artificial intelligence in medical diagnosis

Home Page AI in medical diagnosis What's New Page Contact Page The quality of information G-Lab Members Kalandos fizika Rejtélyes fizika A Gondolkodás Geometriája Több, mint fizika UFO's at G-Lab Custom Blank

About the author

Dr. Gyorfi Andras is a consultant in internal medicine. Graduated in 1986 at the Medical University of Tg. Mures - Marosvasarhely -, is working in the hospital of the city of Marghita - Margitta -, county Bihor in Romania.

The History Behind My Site

I started this project on my cybernetics classes of prof. Laszlo, back in 1985. In the beginnings the decisional algorhytm was built, which was later combined with a database. The modern era came with the module of fuzzy logic for multidimensional paralell data-processing and the database-feedback facility, which brought AI elements to the system.

About Thorax

Thorax is an expert system enhanced with components of artificial intelligence.

Is based on heuristic, deterministic and fuzzy logic.

Is able to run parallell processes in the decisional algorithm, on multiple plans - to determine the probability values of the diseases enrolled to the list of differential dignoses and to explore the effect of the patients subiectivity regarding the intensity of a sign (e.g. pain).

Uses a database with the specifity and sensibility of each simptom and sign for the possible diagnostics, these values allow to calculate the probability of the best diagnostic - and of those enrolled in the differential diagnostics list.

By a process of negative feed-back, after validation, the elicited symptoms and signs are reintroduced to a database wit similar structure in the corespondent diagnostic's column / learning module