Motion Detection System for Recognition of Early Sign of Depression

Depression is one of the mental health disorders that affect many humans, especially campus students. In certain cases, people with depression-prone to commit suicide without any warning signs and symptoms observed by family and friends. There is a need to be able to identify and proceed for treatment from the professionals as soon as possible. There is a lack of tools to identify students' depression behavior through quantified motion characteristics. The advancement of algorithms could be used in detecting such behaviors. This research is motivated to classify depression among students using artificial intelligence. The motion characteristics are quantified using accelerometer and GPS data and trained using neural networks to enable human activity prediction. Once predicted the prone to depression behavior notification will be sent to alert the user on their mental health condition. The user should react and respond to the alert and meet their doctors for further treatment.