Enhancing Anxiety Detection Through ECG-Derived HRV Features and Machine Learning

Anxiety is a mental state characterized by physiological changes such as tense feelings, anxious thoughts, increased blood pressure, and excessive perspiration. Anxiety disorders are common and have a major negative effect on the health of individuals. Detection and timely tracking of anxiety disorders can improve treatment outcomes. The aim of this project is to explore the use of Electrocardiogram (ECG) signals for detecting anxiety. Anxiety and other emotional states influence heart activity, which is reflected in the physiological data provided by ECG signals. Physiological features of Electrocardiogram (ECG) signal quantify the anxiety response in individuals. The Pan-Tompkins algorithm is employed to identify R peaks in the Electrocardiogram (ECG) signal and calculate the R-R intervals, while the Support Vector Machine (SVM) and Logistics Regression (LR) are utilized to classify the data into anxious and non-anxious categories. The findings show that anxiety can be accurately identified from Electrocardiogram (ECG) signals with encouraging accuracy, indicating that there may be potential to develop a non-invasive, impartial tool for anxiety assessment.