Impact of region-based shape alterations in MRI as indicators of Parkinson's Disease.

Authors

Abstract

Parkinson's Disease (PD) is a neuro-degenerative condition that progressively impairs movement, resulting from the loss of dopamine-producing brain cells. Early detection of such a condition helps slowing their progression and allows more effective management of the disease. MRIs are the most common type of imaging technique used for detection of PD without any specific risk to patients. Therefore, in this study, a novel MRI-based method has been proposed to identify structural biomarkers of certain regions within the brain for the detection of PD. The objective of this study is to analyze the alterations in geometrical shapes of the different regions of the brain in PD. The study involves segmentation of six regions of interest in analysis and identification of their geometrical shapes in PD. Several statistical analyses have been performed on the segmented regions to show the most pronounced shape alterations in PD. Further, a geometrical feature-based classification model is subsequently used to validate these findings. The proposed method achieved good accuracy, surpassing several cutting-edge approaches. The results indicate that pinpointing key regions for PD detection may aid early PD detection with low computational cost and reveal subtle brain changes invisible to the human eye.

Tags

Topics