Parkinson's disease detection using sea lion shuffled shepherd optimization assisted deep maxout network

Abstract

Summary When compared to a cloud‐based application, fog computing allows the doctor to make better decisions in an emergency and also helps secure personal data with less latency. Parkinson's disease (PD) is a kind of brain illness, which causes stiffness, shaking, trouble in walking, talking, and so on. There is various detection techniques have been invented by researchers previously, but less accuracy is a major drawback. The main purpose of this research is the effective PD detection by the progression of optimized deep maxout network (DMN), namely, sea lion shuffled shepherd optimization (SLnSSOA)‐based DMN. Here, the developed method involves fog nodes, blockchain (BC), cloud, medical analyzer, and so on. The input data, which comes from the cloud is sent to the preprocessing phase, where data normalization is used to process it. Additionally, SLnSSOA, a combination of the sea lion optimization algorithm (SLnO) and the shuffled shepherd optimization (SSOA), is used to classify diseases using DMN, where the parameters are trained using SLnSSOA. Moreover, this technique provides an improved classification outcome. The developed method attains maximum testing accuracy of 0.908, sensitivity of 0.930, and specificity of 0.875.