A knowledge-based technique for automated detection of ischaemic episodes in long duration electrocardiograms

A novel method for the detection of ischaemic episodes in long duration ECG s is proposed. It includes noise handling, feature extraction, rule-based b eat classification, sliding window classification and ischaemic episode ide ntification, all integrated in a four-stage procedure. It can be executed i n real time and is able to provide explanations for the diagnostic decision s obtained The method was tested on the ESC ST-T database and high scores w ere obtained for both sensitivity and positive predictive accuracy (93.8% a nd 78.5% respectively using aggregate gross statistics, and 90.7% and 80.7% using aggregate average statistics).