ECG waveform analysis by significant point extraction. II. Pattern matching.

From a set of significant points which characterizes the ECG waveform, the pattern matching algorithm detects and classifies QRS complexes. R waves are detected from the analysis of global curvature. Next, the morphology of the QRS complex is determined. QRS complexes with different morphologies are classified by a correlation algorithm. This method is sensitive to changes in shape, such as that of abnormal QRS complexes. The algorithm should be useful in automated analysis of waveforms, such as ECG signals recorded in clinical environments.

[1]  M. L. Ahlstrom,et al.  Digital Filters for Real-Time ECG Signal Processing Using Microprocessors , 1985, IEEE Transactions on Biomedical Engineering.

[2]  W. J. Tompkins,et al.  Estimation of QRS Complex Power Spectra for Design of a QRS Filter , 1984, IEEE Transactions on Biomedical Engineering.

[3]  Ho Soo Lee.  Model-based pattern analysis of signals using frames , 1985 .

[4]  John K. Tsotsos,et al.  CAA: A Knowledge Based System Using Causal Knowledge to Diagnose Cardiac Rhythm Disorders , 1983, IJCAI.

[5]  Kenneth P. Birman,et al.  Rule-Based Learning for More Accurate ECG Analysis , 1982, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  H S Lee,et al.  ECG waveform analysis by significant point extraction. I. Data reduction. , 1987, Computers and biomedical research, an international journal.

[7]  J. R. Cox,et al.  Digital analysis of the electroencephalogram, the blood pressure wave, and the electrocardiogram , 1972 .

[8]  Ivaturi S. N. Murthy,et al.  Syntactic Approach to ECG Rhythm Analysis , 1980, IEEE Transactions on Biomedical Engineering.