Pattern recognition for ECG-monitoring: an interactive method for the classification of ventricular complexes.

Abstract An interactive method is described for the automatic recognition of different types of QRS-complexes for monitoring of CCU-patients. The method can operate on prefiltered single-lead ECG's. After QRS-detection a fiducial point of reference is computed in the ventricular complex. Consecutive QRS-complexes are allocated automatically or by human interaction to clusters, which represent the different types of waveforms. These clusters are defined by the human operator in a learning period. In this way, the patient is used as his own standard of reference. Each cluster has a finite extension, a complex which falls within the boundaries of the cluster is automatically identified as a member of that particular cluster. If not so, human action is requested.

[1]  C A Caceres,et al.  Monitoring of electrocardiograms in a coronary care unit by digital computer. , 1969, JAMA.

[2]  A. C. Young,et al.  Frequency Analysis of the Electrocardiogram , 1960, Circulation research.

[3]  H. Pipberger,et al.  Computer classifications of electrocardiograms. , 1967, Computers and biomedical research, an international journal.

[4]  J. .. Bemmel,et al.  Detection Procedure to Represent the Foetal Heart Rate and Electrocardiogram , 1966 .

[5]  R. Stacy,et al.  Computers in biomedical research , 1967 .

[6]  V. K. Murthy,et al.  On-line real time computer algorithm for monitoring the ECG waveform. , 1970, Computers and biomedical research, an international journal.

[7]  J R Cox,et al.  Some observations on the economics of computer systems for monitoring electrocardiographic rhythms. , 1971, Computers and biomedical research, an international journal.