An Open-Source Feature Extraction Tool for the Analysis of Peripheral Physiological Data

Area of Science:

• Physiological signal processing
• Affective computing
• Digital health

Background:

• Peripheral physiological signals (biosignals) are crucial in psychological studies and healthcare.
• Applications include health tracking, sleep assessment, disease diagnosis, and understanding emotions.

Purpose of the Study:

• To develop a biosignal-specific processing toolbox (Bio-SP tool) for preprocessing and feature extraction.
• To assist researchers in affective computing, digital health, and telemedicine.
• To enable semi-automatic and reliable extraction of physiological patterns.

Main Methods:

• Developed algorithms for signal-specific quality checking, artifact/noise filtering, and segmentation.
• Incorporated features relevant to category discrimination in healthcare.
• Created an open-source MATLAB software with a user-friendly GUI.

Main Results:

• The Bio-SP tool successfully implements state-of-the-art algorithms for biosignal processing.
• Features extracted are relevant for discriminating between disease and non-disease patterns.
• The GUI is compatible with MathWorks Classification Learner for model development.

Conclusions:

• The Bio-SP tool provides a reliable and accessible method for physiological signal analysis.
• It supports researchers in extracting meaningful features for various healthcare applications.
• The open-source nature and GUI facilitate future development and application in digital health and telemedicine.