Efficient classification model for anxiety detection in autism using intelligent search optimization based on Deep CNN

Autism is a brain disease that harmfully impacts a person’s capacity for interpersonal interaction and communication. Autism is also known as autistic spectrum disorder (ASD) because of the vast range of symptoms it exhibits. Anxiety is a common symptom of ASD in children, and it creates unique lifelong barriers that severely limit daily opportunities and negatively impact quality of life. To address the issue, an automatic classification module is proposed for anxiety detection in ASD using intelligent search optimization-based Deep Convolutional Neural Network (CNN). The main contribution of the research rests on intelligent search optimizations (ISO), which successfully tune the classifier's parameters using the fitness function. The accuracy, sensitivity, and specificity of the results are used for evaluating the research's success. The proposed ISO-based Deep CNN attained the values of 95.50%, 94.20%, and 98.49% concerning bandwise performance, 95.83%, 94.59%, 98.81% concerning k-fold and 95.83%, 94.81%, 98.82% concerning training percentage, which shows the effectiveness of the research.