An intelligent system for electrocardiogram diagnosis
Jiguang Ge, Gongyu Shen, Shuoyin Chen, Jinbo Pan · 2002
A primary research real-time system for obtaining a diagnostic description of an electrocardiogram (ECG) sequence has been developed and tested. The knowledge about the ECG has been obtained from highly skilled and experienced physicians and conclusions are drawn by forward reasoning and refinement into an heuristic approach based on rules. On the low level, the ECG parameters are extracted and patterns are recognized; these data are the input for the high level processing. Tests with 50 rules gave correct descriptions as compared to the diagnosis of a physician. The coincident rate of diagnostic results is over 90 percent.