Mental Stress Assessment Using Ultra Short Term HRV Analysis Based on Non-Linear Method.

Area of Science:

• Physiology
• Biomedical Engineering
• Data Science

Background:

• Mental stress is a growing public health concern linked to various diseases.
• Early detection and management of stress are vital for maintaining well-being.
• Traditional stress detection methods may lack real-time applicability.

Purpose of the Study:

• To introduce novel heart rate variability (HRV) features derived from empirical mode decomposition.
• To investigate the efficacy of short-term (5 min) and ultra-short-term (under 5 min) HRV analysis for acute mental stress detection.
• To identify optimal HRV analysis durations for practical stress monitoring applications.

Main Methods:

• Acquired HRV signals from 74 young police officers subjected to acute stressors (Trier Social Stress Test, horror movies).
• Extracted 26 features, including novel Intrinsic Mode Function (IMF) energy features and standard HRV metrics.
• Employed a Support Vector Machine (SVM) classifier with leave-one-subject-out cross-validation for stress/non-stress classification.

Main Results:

• Ultra-short-term HRV analysis achieved a classification accuracy of 90.5%.
• Short-term HRV analysis demonstrated a classification accuracy of 86.5%.
• Analysis identified optimal durations for ultra-short-term HRV, suitable for wearable and healthcare systems.

Conclusions:

• Ultra-short-term HRV analysis, utilizing novel features, is highly effective for detecting acute mental stress.
• The findings support the integration of HRV-based stress detection into wearable devices and healthcare systems.
• This research offers a promising, non-invasive approach to proactive mental stress management.