Prediction of Activity pattern mining for Neurological disease using Convolution Neural Network

The brain, spinal cord, and nerves make up the nervous system in the human body. They are in charge of the entire body’s operation when they function together. If any portion of the neurological system in the body malfunctions, it will impact the rest of one’s life and activities. We cannot perform usual activities because of disruptions in our nervous systems. A neurological condition characterized by poor coordination of movement and balance difficulties is known as Activity Pattern Mining (APM). The clinical procedure is critical since it takes longer to diagnose and treat the disease. This research examines the potential of identifying neurological disorders, especially Cerebellar ataxia (CA) using noninvasive treatment activity pattern mining. The sensors are used on the body to collect the data and record limb movement. The heel-knee-movement test identifies the patient’s essential activity pattern motions. In this research, we experimented with the CA dataset and applied Recurrent Neural Network (RNN) and Convolution Neural Network (CNN). The least-squares approach and accuracy of RNN and CNN algorithms were employed to optimize the training and the prediction model is built to predict the brain disease functions. The binary classification of the model will analyze and predict the neurological disease.