Parkinson’s disease is the second most prevalent neurological disease, affecting millions of people globally. It is a condition that affects different regions of the brain in the basal ganglia, which is characterized by motor symptoms and postural instability. Currently, there is no cure available in order to completely eradicate the disease from the body. As a result, early diagnosis of Parkinson’s Disease (PD) is critical in combating the gradual loss of dopaminergic neurons in patients. Although much progress has been made in using medical images such as MRI and DaTScan for diagnosing the early stages of Parkinson’s Disease, the work remains difficult due to lack of properly labeled data, high error rates in clinical diagnosis and a lack of automatic detection and segmentation software. In this paper, we propose a software called PPDS (Parkinson’s Disease Diagnosis Software) for the detection and segmentation of deep brain structures from MRI and DaTScan images related to Parkinson’s disease. The proposed method utilizes state-of-the-art convolutional neural networks such as YOLO and UNET to correctly identify and segment regions of interest for Parkinson’s disease from both DatScan and MRI images, as well as deliver prediction results. The aim of this study is to evaluate the performance of deep convolutional networks in automating the task of identifying and segmenting the substantia nigra and striatum from T2-weighted MRI and DatScan images respectively, which are used to monitor the loss of dopaminergic neurons in these areas.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.