Deep Learning-Enabled Clinically Applicable CT Planbox for Stroke With High Accuracy and Repeatability

• Published in: Frontiers in neurology Vol. 13; p. 755492
• Main Authors: Wang, Yang, Zhu, Junkai, Zhao, Jinli, Li, Wenyi, Zhang, Xin, Meng, Xiaolin, Chen, Taige, Li, Ming, Ye, Meiping, Hu, Renfang, Dou, Shidan, Hao, Huayin, Zhao, Xiaofen, Wu, Xiaoming, Hu, Wei, Li, Cheng, Fan, Xiaole, Jiang, Liyun, Lu, Xiaofan, Yan, Fangrong
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Media S.A 11.03.2022
• Subjects: accurate and repeatable images; automatic cranial scanning; computed tomography; deep learning; Neurology; stroke; computed tomography; automatic cranial scanning; deep learning; stroke; accurate and repeatable images
• ISSN: 1664-2295; 1664-2295
• DOI: 10.3389/fneur.2022.755492

Computed tomography (CT) plays an essential role in classifying stroke, quantifying penumbra size and supporting stroke-relevant radiomics studies. However, it is difficult to acquire standard, accurate and repeatable images during follow-up. Therefore, we invented an intelligent CT to evaluate stroke during the entire follow-up. We deployed a region proposal network (RPN) and V-Net to endow traditional CT with intelligence. Specifically, facial detection was accomplished by identifying adjacent jaw positions through training and testing an RPN on 76,382 human faces using a preinstalled 2-dimensional camera; two regions of interest (ROIs) were segmented by V-Net on another training set with 295 subjects, and the moving distance of scanning couch was calculated based on a pre-generated calibration table. Multiple cohorts including 1,124 patients were used for performance validation under three clinical scenarios. Cranial Automatic Planbox Imaging Towards AmeLiorating neuroscience (CAPITAL)-CT was invented. RPN model had an error distance of 4.46 ± 0.02 pixels with a success rate of 98.7% in the training set and 100% with 2.23 ± 0.10 pixels in the testing set. V-Net-derived segmentation maintained a clinically tolerable distance error, within 3 mm on average, and all lines presented with a tolerable angle error, within 3° on average in all boundaries. Real-time, accurate, and repeatable automatic scanning was accomplished with and a lower radiation exposure dose (all < 0.001). CAPITAL-CT generated standard and reproducible images that could simplify the work of radiologists, which would be of great help in the follow-up of stroke patients and in multifield research in neuroscience.