Classification Method of Children’s Attention Deficit Hyperactivity Disorder Based on Double-Layer Twin Neural Network

In order to accurately classify children’s attention deficit hyperactivity disorder types and improve the accuracy and reliability of classification, a classification method of children’s attention deficit hyperactivity disorder types based on double-layer twin neural network is proposed. Divide the characteristics of children’s attention deficit hyperactivity disorder into impulse/hyperactivity type, attention deficit type and mixed type, establish a classification formula, calculate the classification state domain and get the classification target center result. The forward learning method and backward learning method of the double-layer twin neural network are designed, and the structure of the double-layer twin neural network is established. The results of the classification of obstacle types are used as the input of the double-layer twin neural network, and finally the classification of children’s attention deficit hyperactivity disorder types is realized. The experimental results show that the classification error of this method is small, which is less than 15% and the classification accuracy is higher, which is more than 95%. The classification results of children’s attention deficit hyperactivity disorder types are completely consistent with the original text, which has good adaptability and efficiency.