CLASSIFICATION OF SEVERITY OF MEDICAL CONDITION BY WAVELET BASED MULTI-RESOLUTION ANALYSIS

摘要

A system and method are provided for classifying the severity of a medical condition detected by a medical device using wavelet based multi-resolution analysis of biomedical signals. Biomedical signals are converted into a set of wavelet coefficients including scales of wavelet coefficients at different resolutions ranging from a finest scale to a coarsest scale. Features of the biomedical signal useful in diagnosing a biomedical condition are identified by determining whether corresponding identifying features appear in at least one of the plurality of scales of wavelet coefficients. The identifying features determined to appear in the various scales of wavelet coefficients are conditionally linked together and used to classify a degree of severity of a diagnosed biomedical condition based on the number of identifying features appearing in different scales of wavelet coefficients. In some embodiments, alerts can be generated based on the diagnosed degree of severity of the biomedical condition.