Case-based learning for prediction of post-myocardial infarction outcomes

It is of prime importance to ascertain the prognosis of patients who have experienced heart attack or unstable angina, as they are prone to developing serious adverse after-effects. The aim of this study was to run machine learning algorithms over a cardiac database to derive associations between various clinical and pathological parameters and the occurrence of future adverse consequences. The rules induced from the data were used to build an expert system for prediction of outcomes for unseen cases and it was ported on the web for use over the Internet.