Early Detection of Anxiety, Depression and Stress among Potential Patients using machine learning and deep learning models

Abstract

Depression, anxiety and stress are the three important mental health conditions. Only a small portion of the millions of people who experience depression, anxiety and stress each year receives a prompt treatment. The development of technology that helps clinicians detect various types of mental diseases has greatly benefited from the progress of Machine Learning (ML) techniques over the past few decades. This study presents a comparison of three important mental health issues (Depression, Anxiety, and stress) using machine learning and Deep Learning (DL) algorithms. The proposed work also investigates about how the ML and DL techniques might help with the detection, diagnosis, and management of mental health diseases with a better effective and novel approach. As a result, three different algorithms (SVM, ANN and XgBoost) has been implemented in which Support vector machine (SVM) has achieved the better accuracy in detecting the mental health. The accuracy achieved in detecting depression, anxiety and stress are 99.32%, 99.80% and 98.44% respectively. Henceforth the proposed study was able to achieve a satisfactory result that helps to improve the decision-making and clinical practice.