Heart rate variability metrics for fine-grained stress level assessment

Area of Science:

• Biomedical Engineering
• Physiological Monitoring
• Autonomic Nervous System Research

Background:

• Accurate, fine-grained stress assessment remains a challenge in biomedical engineering.
• Short-term stress monitoring overcomes limitations of long-term analysis, enabling real-world behavior and stress dynamics evaluation.
• Heart rate variability (HRV) analysis is a key physiological index for stress estimation.

Purpose of the Study:

• To evaluate time, frequency, and nonlinear HRV metrics for short-term stress level assessment.
• To identify reliable HRV metrics for continuous stress monitoring.
• To determine the optimal window length for stress detection using HRV.

Main Methods:

• Electrocardiogram (ECG) signals were recorded from 14 volunteers using Vital Jacket™.
• The Trier Social Stress Test (TSST) protocol was employed to induce standardized stress.
• HRV analysis was performed using varying window lengths, from 220s down to 50s.

Main Results:

• A subset of HRV metrics, including AVNN, rMSSD, SDNN, and pNN20, demonstrated consistent differences between stress and non-stress conditions.
• These metrics proved reliable for short-term stress level assessment.
• The AVNN metric with a 50s window length showed significant reliability in recognizing stress across TSST phases.

Conclusions:

• The AVNN metric, analyzed with a 50s window, is the most reliable for recognizing stress levels throughout the TSST.
• This approach enables a fine-grained analysis of stress effects and provides insight into autonomic nervous system responses.
• Short-term HRV analysis, particularly using AVNN, is a promising tool for assessing psychological stress.