AI-Based Stress and Anxiety Detection System Using Ayurvedic Parameters and Machine Learning Techniques

Abstract

Stress and anxiety disorders have emerged as major global health concerns due to increasing work pressure, lifestyle imbalance, emotional instability, and digital dependency. Traditional diagnostic systems primarily rely on psychological evaluation and clinical observations, often overlooking personalized and holistic health indicators. Ayurveda, the ancient Indian system of medicine, emphasizes the relationship between mental wellness, lifestyle, diet, sleep, and the balance of Tridoshas (Vata, Pitta, and Kapha). This research proposes an Artificial Intelligence (AI)-based stress and anxiety detection framework integrating Ayurvedic principles with machine learning techniques for early diagnosis and personalized wellness recommendations. The proposed system utilizes questionnaire-based data collection including behavioral, psychological, and Ayurvedic parameters. Multiple machine learning algorithms such as Random Forest, Support Vector Machine (SVM), Decision Tree, Logistic Regression, and XGBoost are analyzed for predictive performance. The study highlights literature review, methodology, system architecture, dataset preparation, model training, evaluation metrics, advantages, limitations, ethical considerations, and future scope. The proposed framework demonstrates the potential of combining AI and Ayurveda to create explainable, preventive, and personalized mental healthcare systems.