همایش بین المللی هوش مصنوعی

صفحه اصلی / اولین همایش بین المللی هوش مصنوعی

Evaluating Parkinson’s Disease Severity Through Attention-Based STGCN and S2AGCN Models Utilizing Kinect Skeleton Images

نویسندگان :

Fatemeh Fadaei Ardestani¹ Nima Asadi²

1- university of Isfahan 2- University of Maryland

کلمات کلیدی :

3D Motion tracking،Computational neurology،Deep learning diagnostics،Motor function analysis،Parkinson's assessment

چکیده :

Parkinson's Disease (PD) is a prevalent neurological disorder marked by motor symptoms such as rigidity and tremors. Accurate and timely assessment of disease severity is essential for judging the efficacy of various treatment interventions. This study presents an innovative approach that employs computer vision technology—specifically Kinect cameras—paired with advanced deep learning techniques to enable precise evaluations of Parkinson's severity. Leveraging the high accuracy of Kinect cameras in capturing critical movement patterns, our proposed system utilizes advanced convolutional neural networks, including the mechanisms used by Spatial-Temporal Graph Convolutional Network (STGCN) and the Two-Stream Adaptive Graph Convolutional Network (2SAGCN). These networks are designed to detect movement anomalies and produce quantitative severity measures effectively. Additionally, we propose architectural enhancements to the 2SAGCN that integrates channel and graph attention modules, resulting in improved classification performance. The severity classification framework distinguishes between 11 specific classes of Parkinson's symptoms, which are derived from 9 distinct motion categories. Within this framework, class 0 represents healthy individuals, while classes 0.1 to 1 correspond to varying degrees of severity in Parkinson's symptoms, resulting in a comprehensive classification system encompassing 99 distinct outcomes. To further enhance the model’s accuracy, we have implemented strategies such as three-dimensional transfer learning and data augmentation. This research marks a significant step forward in the realm of non-invasive, quantitative assessments of Parkinson's Disease, showcasing the potential of cutting-edge technology and state-of-the-art neural network architectures.