OMAD: On-device Mental Anomaly Detection for Substance and Non-Substance Users

TL;DR

This paper presents OMAD, a real-time on-device system using EEG signals from wearable devices to detect mental health anomalies related to substance use, with optimized models suitable for resource-constrained hardware.

Contribution

It introduces a novel artifact removal and activity recognition framework optimized for wearable devices, enabling real-time mental health anomaly detection on low-power hardware.

Findings

Achieved approximately 93% accuracy in artifact removal

Achieved approximately 90% accuracy in activity detection

Reduced model size by 70% and inference time by 31%

Abstract

Stay at home order during the COVID-19 helps flatten the curve but ironically, instigate mental health problems among the people who have Substance Use Disorders. Measuring the electrical activity signals in brain using off-the-shelf consumer wearable devices such as smart wristwatch and mapping them in real time to underlying mood, behavioral and emotional changes play striking roles in postulating mental health anomalies. In this work, we propose to implement a wearable, {\it On-device Mental Anomaly Detection (OMAD)} system to detect anomalous behaviors and activities that render to mental health problems and help clinicians to design effective intervention strategies. We propose an intrinsic artifact removal model on Electroencephalogram (EEG) signal to better correlate the fine-grained behavioral changes. We design model compression technique on the artifact removal and activity recognition (main) modules. We implement a magnitude-based weight pruning technique both on convolutional neural network and Multilayer Perceptron to employ the inference phase on Nvidia Jetson Nano; one of the tightest resource-constrained devices for wearables. We experimented with three different combinations of feature extractions and artifact removal approaches. We evaluate the performance of {\it OMAD} in terms of accuracy, F1 score, memory usage and running time for both unpruned and compressed models using EEG data from both control and treatment (alcoholic) groups for different object recognition tasks. Our artifact removal model and main activity detection model achieved about $\approx$ 93\% and 90\% accuracy, respectively with significant reduction in model size (70\%) and inference time (31\%).