Design and Implementation of Portable Abnormal ECG Signal Analysis Instrument Based on Feature Classifcation

Abstract

To collect and analyze the ECG signal in real time, the analog filter and the signal amplifier were used to construct the abnormal signal acquisition and classification system. The ARM10E processor was used to detect the signal shape and QRS complex wave. Based on the Poincare support vector machine, the feature set was extracted from the training data set to construct the heart disease classifier, and the clinical classification model was given. The device effectively reduces computational complexity, improves processor speed, real-time acquisition and diagnoses heart disease. Portable ECG devices can capture suspected waveforms of abnormal signals, establish and evaluate high quality signals, reduce patient on-line waiting time, and facilitate early diagnosis and recognition of heart disease.