Search

Zhiliang Zhang,1,&ast; Zhongxiang Ding,2,&ast; Fenyang Chen,2 Rui Hua,3 Jiaojiao Wu,3 Zhefan Shen,2 Feng Shi,3 Xiufang Xu1 1School of Medical Imaging, Hangzhou Medical College, Hangzhou, People’s Republic of China; 2Department of Radiology, Affiliated Hangzhou First People’s Hospital, Westlake University School of Medicine, Hangzhou, People’s Republic of China; 3Department of Research and Development, Shanghai United Imaging Intelligence Co., Ltd, Shanghai, People’s Republic of China&ast;These authors contributed equally to this workCorrespondence: Feng Shi; Xiufang Xu, Email feng.shi@uii-ai.com; 1996033011@hmc.edu.cnBackground: Cerebral small vessel disease lacks specific clinical manifestations, and extraction of valuable features from multimodal images is expected to improve its diagnostic accuracy. In this study, we used deep learning techniques to segment cerebral small vessel disease imaging markers in multimodal magnetic resonance images and analyze them with clinical risk factors.Methods and results: We recruited 211 lacunar stroke patients and 83 control patients. The patients’ cerebral small vessel disease markers were automatically segmented using a V-shaped bottleneck network, and the number and volume were calculated after manual correction. The segmentation results of the V-shaped bottleneck network for white matter hyperintensity and recent small subcortical infarction were in high agreement with the ground truth (DSC> 0.90). In small lesion segmentation, cerebral microbleed (average recall=0.778; average precision=0.758) and perivascular spaces (average recall=0.953; average precision=0.923) were superior to lacunar infarct (average recall=0.339; average precision=0.432) in recall and precision. Binary logistic regression analysis showed that age, systolic blood pressure, and total cerebral small vessel disease load score were independent risk factors for lacunar stroke (P< 0.05). Ordered logistic regression analysis showe