Machine learning for biomedical signal processing

Among the various applications of machine learning in Biomedical Engineering, one of the ares of focus for researchers is its application in biomedical signal processing to extract, analyze, and categorize various signals or images for diagnosis purposes. The biomedical signals are mainly acquired in nonlinear and time-varying environments. Machine learning can also be applied effectively for the processing and classification of bioelectric signals, like an electrocardiogram (ECG), electroencephalogram (EEG), electromyogram (EMG), etc. Processing of an ECG signal can be categorized into three stages: preprocessing (filtering), feature extraction, and classification. This chapter focuses on the application of machine learning–based algorithms at various stages of processing of an ECG signal to extract useful information for early and accurate detection of cardiac disorders in order to provide effective treatment for patients.