A novel physiological feature selection method for emotional stress assessment based on emotional state transition

• DOI: 10.3389/fnins.2023.1138091
• Publication Date: 2023-03-22T05:10:38Z
• Authors: Zhen Li; Yun Xing; Yao Pi; Mingzhe Jiang; Lejun Zhang
• Keywords: Feature (linguistics); Emotional stress

The connection between emotional states and physical health has attracted widespread attention. stress assessment can help healthcare professionals figure out the patient's engagement toward diagnostic plan optimize rehabilitation program as feedback. It is of great significance to study changes physiological features in process change find subset one or several that best represent psychological state a statistical sense. Previous studies had used differences discrete select feature subsets. However, human body continuously changing. conventional selection methods ignored dynamic an individual's real life. Therefore, dedicated experimental was conducted while three peripheral signals, i.e., ElectroCardioGram (ECG), Galvanic Skin Resistance (GSR), Blood Volume Pulse (BVP), were acquired. This paper reported novel method based on transition, results show number selected by proposed this 13, including 5 ECG, 4 PPG GSR, respectively, which are superior PCA terms dimension reduction. classification accuracy emotion recognition ECG higher than other two methods. These suggest serve viable alternative methods, transition deserves more attention promote development assessment.