Detection and Classification of Parkinson disease using various features extraction model and deep Learning Techniques for early stages

Parkinson's disease (PD) is a neurological movement disorder characterized by a slow, progressive worsening of symptoms, such as a tremor in one hand and a generalized sense of stiffness. More than 6 million individuals throughout the globe are afflicted. In the early stages of the illness, when symptoms are hard to identify, there is currently no convincing finding for this condition by non-specialist practitioners. To better understand patients, an RNN-based predictive analytics system is developed. The issue can be resolved with a small margin of error utilizing deep learning methods. If input data sets are to be used for analysis, they should be retrieved from the UCI Machine Learning repository. The goal of this research is to create an auditory feature-based method for detecting Parkinson's disease. Several machine learning methods are used to model the retrieved characteristics. In this study, we apply an RNN-based classification technique to identify PD patients' samples from those of healthy individuals. The RNN network is taught acoustic characteristics and a spectrogram of the speaker's voice. Only auditory characteristics are used in the training of RNN models. Using optimization strategies, Deep Learning (RNN) algorithms, and health care data, this research establishes whether or not people have Parkinson's disease. To test the efficiency and performance of the proposed method, a comparative research is conducted.