Deep-learning methods (especially convolutional neural networks) using magnetic resonance imaging (MRI) data have been successfully applied to computer-aided diagnosis of Parkinson's Disease (PD). Early detection and prior care may help patients improve their quality of life, although this neurodegenerative disease has no known cure. In this study, we propose a FResnet18 model to classify MRI images of PD and Health Control (HC) by fusing image texture features with deep features. First, Local Binary Pattern and Gray-Level Co-occurrence Matrix are used to extract the handcrafted features. Second, the modified ResNet18 network is used to extract deep features. Finally, the fused features are classified by Support Vector Machine. The classification accuracy rate for MRI images reaches 98.66%, and the findings demonstrate that the model can successfully differentiate between PD and HC. The suggested FResnet18 provides greater performance compared with existing approaches, and it is shown through extensive experimental findings on the Parkinson's Disease Progression Markers Initiative data set that feature fusion may improve classification performance.

中文翻译：

---

基于T2 MRI图像特征融合的帕金森病诊断

使用磁共振成像 (MRI) 数据的深度学习方法（尤其是卷积神经网络）已成功应用于帕金森病 (PD) 的计算机辅助诊断。早期发现和预先护理可能有助于患者改善他们的生活质量，尽管这种神经退行性疾病尚无已知的治愈方法。在这项研究中，我们提出了一个 FResnet18 模型，通过融合图像纹理特征和深层特征来对 PD 和健康控制 (HC) 的 MRI 图像进行分类。首先，使用局部二进制模式和灰度共生矩阵来提取手工制作的特征。其次，修改后的 ResNet18 网络用于提取深度特征。最后，融合后的特征通过支持向量机进行分类。MRI图像分类准确率达到98.66%，结果表明，该模型可以成功地区分 PD 和 HC。与现有方法相比，建议的 FResnet18 提供了更高的性能，并且通过对帕金森病进展标记计划数据集的大量实验结果表明，特征融合可以提高分类性能。