AI-Driven Integrated Hardware and Software Solution for EEG-Based Detection of Depression and Anxiety

Depression and anxiety are prevalent mental disorders that have impacted a substantial number of individuals worldwide, exceeding 300 million cases. The repercussions of the COVID-19 pandemic are expected to further escalate these figures due to the economic, social, and societal challenges faced by individuals. Extensive research has revealed distinctive asymmetry in frontal brain-wave activity among individuals with depression and anxiety compared to those without these disorders. Considering this, our research proposes a non-invasive method utilizing a wearable EEG device for the detection of depression and anxiety. The study encompasses essential components such as EEG device interfacing, data collection, preprocessing, feature extraction, data analysis, machine learning model training and evaluation, and the development of a mobile application enabling on-device inference and integration with a cloud database. EEG signals were collected from 30 individuals in a resting state using a single-electrode EEG sensor. Time and frequency domain analyses were conducted on the collected signals. Our machine learning model achieved a remarkable 93% accuracy in detecting depression and anxiety. Thus, the completed study comprises both hardware and software elements. The hardware component features the NeuroSky Mindwave wearable EEG sensor, while the software component includes machine learning models, an Android mobile application, and a data processing pipeline. This integrated system aims to provide a comprehensive solution for the detection and management of depression and anxiety, ultimately enhancing the well-being of individuals afflicted by these conditions.