A Hospital Application Involving Deep Learning Architecture for Detecting Parkinson’s Disease

Parkinson’s disease is a nervous system disease that affects a person’s ability to control movement. The main reason for the occurrence of the disease is due to the abnormal loss of neurons (brain cells) in the brain. Symptoms involve uncontrollable tremors in the hand, muscle stiffness and have trouble maintaining balance. The most predominant signs and symptoms confirming the start of the disease are tremors, Medical diagnosis of the Parkinson’s disease involve verification of Magnetic Resonance Imaging scans of brain and various other lab test by doctors. This manual interpretation of medical images demands high time consumption and is highly prone to mistakes. In this project modified VGG Net architecture is applied to accurately detect the Parkinson’s disease from the Magnetic Resonance Imaging scans of a patient’s presence without the need of several consultations from different doctors. This leads to earlier and accurate detection of the disease and allows us to take prior actions immediately to avoid further worsening of the symptoms in an effective and cheap manner avoiding human error rate. A web application using a JavaScript framework reactJS will be developed as a front-end interface for the project. Thus, this project is used to reduce the human error rate for medical diagnosis of the Parkinson’s disease effectively.