Deep Learning-based Early Parkinson’s Disease Detection from Brain MRI Image

Recent decade, Parkinson’s disease (PD),which impairs the life quality for millions of older people worldwide, has quickly emerged as a serious condition affecting the brain and spinal cord. Appropriate treatment and management of the disease depend on early discovery and an accurate diagnosis. Due to PD’s close resemblance to other neurological disorders, the precise diagnosis of PD has until now been difficult. These same characteristics account for 25% of incorrect manual PD diagnosis. Brain MRI (Magnetic Resonance Imaging) has shown great potential in the detection and diagnosis of Parkinson’s disease. Proposed study uses convolutional neural networks (CNN), a type of deep neural network architecture, to classify Parkinson disease in order to differentiate between PD patients and healthy controls. Parkinson Progression Markers Initiative (PPMI)dataset is used as input to classify the disease. Here, the median filtering technique is used to remove the noise from the images and preserve the edges which help to provide a letter image and able to predict it easily. The Parkinson disease recognition system is done by using CNN. Accuracy, sensitivity, specificity, and AUC (Area Under Curve) are used to assess the performance of the suggested approach.