A biosignal-specific processing tool for machine learning and pattern recognition

Electrocardiogram (ECG), Electrodermal Activity (EDA), Electromyogram (EMG) and Impedance Cardiography (ICG) are among physiological signals widely used in various biomedical applications including health tracking, sleep quality assessment, early disease detection/diagnosis and human affective state recognition. This paper presents the development of a biosignal-specific processing and feature extraction tool for analyzing these physiological signals according to the state-of-the-art studies reported in the scientific literature. This tool is intended to assist researchers in machine learning and pattern recognition to extract feature matrix from these bio-signals automatically and reliably. In this paper, we provided the algorithms used for the signal-specific filtering and segmentation as well as extracting features that have been shown highly relevant to a better category discrimination in an intended application. This tool is an open-source software written in MATLAB and made compatible with MathWorks Classification Learner app for further classification purposes such as model training, cross-validation scheme farming, and classification result computation.