Designing a Smart and Intelligent Ecosystem for Autistic Children
Maddhuja Sen, Kaustav Sharma, Sushruta Mishra, Ahmed Alkhayyat, Vandana Sharma
Abstract
In this paper, we discuss the usage of smart and machine learning based model to aid children with autism. Children with autism require constant care because you never know what will trigger them. A novel method is suggested in this paper where the child is constantly under CCTV camera when he or she is alone in a room. The child's movements will be watched on camera. Data from the camera will be gathered and fed into a machine learning model, which will then be able to anticipate whether or not that particular youngster has been triggered. According to the prediction, the caregiver will be notified and will be updated with ways to calm the child down. The youngster will be wearing a smartwatch that, in the event of a caretaker delay in arrival, will detect a spike in heart rate and momentarily calm the child by playing calming music. The autistic child will benefit much from this approach, and it will also make it easier for the caregiver to track the child's activities using a mobile application. Random forest records the best performance with an accuracy of 94.7% and least response delay of 3.44 seconds.