Abstract

In this paper, a data acquisition system for QRS detections in 12-lead electrocardiogram (ECG) diagnosis is proposed. This system collect data, learns rules, synthesizing decision trees by inductive inference from examples. An extended form of Quinlan's algorithm is used. There is also possibility of continual improving the knowledge base maintained, by learning new rules and merging them with the old ones. As a start point for the knowledge base learned, the rules of an already developed expert system are used.

Authors and Affiliations