Non-EEG Method To Predict a Psychiatric Disorder and Proposed Preventive Method

Abstract

Social anxiety disorder (SAD) involves an intense fear of social interactions, leading to distress and impaired daily functioning. This study aims to develop a wearable technology to predict and mitigate anxiety attacks in SAD patients using Photoplethysmogram (PPG) and Electrocardiogram (ECG) signals. The system analyzes data from 135 participants, using the Liebowitz scale and State-Trait Anxiety Inventory to select 30 individuals for detailed analysis. Key features from ECG and PPG signals were input into a Naive Bayes algorithm, achieving an 84.24% accuracy in predicting anxiety states. The system also provides sound and vibration stimuli to help calm patients, potentially improving their quality of life.