Mental health status monitoring for people with autism spectrum disorder using machine learning

In this article, an assistive technology in the form of a stress monitoring system customized exclusively for people affected by autism spectrum disorder (ASD) is proposed. ASD tends to impact the linguistic and motor skills and the effects of stress often has a negative impact on the physical and mental health of the individuals. The suppressed heart rate variability (HRV) is found to be lesser in ASD affected individuals. This is considered as a key factor and the constraints of the parameters to be monitored such as heart rate, body temperature and body movements is set accordingly. The hardware module encompasses various sensors such as pulse sensor, temperature sensor and accelerometers and can be developed into a variety of wearables specific to the user. The acquired data is transmitted to a non-relational database whose data is extracted and processed by a machine learning model consisting of a linear regression algorithm and random forest classifier algorithm to generate and analyse reports to support the mental stress management of the user. This wearable device can be accommodated for remote monitoring of the users in mundane day to day situations and even in social habitats such as healthcare establishments and even in educational institutions. © 2023, The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of Computer Applications and Management.