Heart Rate Variability and Machine Learning for Stress Analysis

Stress is a feeling of being overwhelmed or unable to cope with emotional or mental pressure. An individual under stress can have physical symptoms, such as high blood pressure, chest pain, headaches and it can also lead to panic attacks, depression, and other forms of anxiety. In this study, we propose an algorithm for the classification of stress levels using the frequency domain parameters of heart rate variability (HRV). The power spectral density of the electrocardiogram (ECG) signals were studied using Welch's method. Further, a machine learning model based on the K-Nearest Neighbours (KNN) classifier was used to analyze varying stress levels in a subject. An F1-score of 0.9 was achieved for stress.