A Fuzzy Expert System for Prognosis of the Risk of Development of Heart Disease



Department of Artificial Intelligence, Shahr-e-Qods Branch, Islamic Azad University, Tehran, Iran


Fuzzy logic has a high potential for managing the uncertainty sources associated with the medical expert systems. Application of fuzzy inference model has been widely concentrated for managing uncertainties in computer based practices of medicine. This paper has proposed two fuzzy expert systems for prognosis of the heart disease based on: 1) Mamdani inference model, and 2) Sugeno inference model. These methods initially received clinical parameters as input sand define their corresponding fuzzy sets. The performance of the FESs (Fuzzy Expert System) based on the Mamdani and Sugeno model, have been evaluated using real patients dataset through conducting two different studies. The dataset includes 380 real cases collected from the Parsian Hospital in Karaj. The accuracy of the proposed Mamdani FES is equal to79.47% and its accuracy using Sugeno model is equal to 88.43%. This FES is promising for prognosis of the heart disease and consequently early diagnosis of the disease and improving survival rates.