Computer-Aided Diagnosis of Parkinson’s Disease Based on Dual
Mode Transcranial Sonography
YAN Minjun1
, WANG Wengang2
, XU Wenwei1
1. Department of Logistics Support, Tongren Hostpital, Shanghai Jiaotong University School of Medicine, Shanghai 200336, China;
2. Hospital Administration Office, Shanghai First Maternity and Infant Hospital, Shanghai 201204, China
Abstract:Objective In order to construct a computer-aided diagnosis system for Parkinson’s disease based on the fusion of
transcranial B-mode sonography (TBS) and transcranial Doppler sonography (TDS), this paper not only exploit the characteristics
of Parkinson’s disease (PD) contained in TBS and TDS, but also make use of the the complementary discriminant information
between the two modalities. Methods The statistical characteristics and texture features were extracted from the TBS images, and
the characteristics of the bleeding flow were calculated from the spectral curves of the TDS images. Then, the bimodal image features
were input into a board learning system (BLS), which was used to classify PD. Results The experimental results show that the fusion
of dual mode transcranial sonography (TCS) imaging methods was better, and the performance of the classifier using BLS was better
than that of the SVM classifier. Conclusion The results can help us to understand the characteristics of TCS in PD and to realize the
automatic diagnosis of PD.
颜敏俊1
,王文刚2
,徐文蔚1. 基于双模经颅超声成像的帕金森病计算机辅助诊断研究[J]. 中国医疗设备, 2020, 35(5): 109-113.
YAN Minjun1
, WANG Wengang2
, XU Wenwei1. Computer-Aided Diagnosis of Parkinson’s Disease Based on Dual
Mode Transcranial Sonography. China Medical Devices, 2020, 35(5): 109-113.