An Innovative Approach to Predict early Stage Depression using Mobile Network Usage

Abstract

This article introduces an early-stage depression identification technique without a physical presence in the medical unit. Currently used methods do not provide any early-stage depression detection techniques and require continuous investigation to diagnose depression. In general, depression detection is based on changes in human behavior. Changes in human behavior are also reflected in his/her mobile usage pattern. Mobile usage can correlate with human behavior and predict early-stage depression based on artificial intelligence (AI). The mobile usage data collected (e.g. location, activity, phone usage) can be used for effective depression screening. The proposed approach suggests using the user’s mobile usage pattern and applying machine learning algorithms to identify the change in mobile usage patterns over time. The machine learning algorithm like Density-Based Spatial Clustering of Applications with Noise (DBSCAN) can be effectively used to identify the change in pattern. This change in mobile usage pattern is then used to detect if the user is slowly moving into depression and warn the individual or concerned authority to handle the case.