A Comprehensive Approach to Predict the Autism Spectrum Disorders using Deep Learning Mechanism

Children often suffer from Autism Spectrum Disorder (ASD), a neurological disease. With stereotypical behavior and difficulties in social communication, ASD cannot be defined by chronological age alone. It is speculated that this condition results from malfunction in the brain caused by the interplay of several ecological, inherited and regulatory variables. More than one percent of the world's population has ASD, prompting policymakers, educators, and health care professionals to develop plans to combat the condition. Maximizing the potential of patients with autism demands a multidisciplinary approach. Therefore, it is crucial that various care models complement one another and do not cause the patient any unnecessary anxiety. Natural therapies have the potential to significantly enhance the long-term success of conventional treatments including psychotherapy, Occupational-Therapy (OT) and PT. According to the progression of symptoms, ASD can take on a number of different forms. However, there is still no agreed-upon gold standard therapy for ASD. If a systematic method is adopted, assessing a kid with suspected ASD is straightforward. There are substantial gains in interpersonal interaction, replication, as well as intellectual and adaptive performance following initial treatment. In this paper, a novel deep learning based prediction approach is designed called as Intelligent Learning Scheme to Predict Autism (ILSPA). This algorithm is cross-validated with the conventional learning approach called Deep Neural Network (DNN) to evaluate the performance and efficiency of the proposed scheme. The resulting scenario provides the proper proof to the efficiency of the proposed algorithm called ILSPA.