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

Yang Wang; Junkai Zhu; Jinli Zhao; Wenyi Li; Xin Zhang; Xiaolin Meng; Taige Chen; Ming Li; Meiping Ye; Renfang Hu; Shidan Dou; Huayin Hao; Xiaofen Zhao; Xiaoming Wu; Wei Hu; Cheng Li; Xiaole Fan; Liyun Jiang; Xiaofan Lu; Fangrong Yan

doi:10.3389/fneur.2022.755492

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

Front Neurol. 2022 Mar 11:13:755492. doi: 10.3389/fneur.2022.755492. eCollection 2022.

Authors

Yang Wang¹, Junkai Zhu², Jinli Zhao³, Wenyi Li⁴, Xin Zhang⁵, Xiaolin Meng⁶, Taige Chen⁷, Ming Li¹, Meiping Ye¹, Renfang Hu⁸, Shidan Dou⁶, Huayin Hao⁶, Xiaofen Zhao⁹, Xiaoming Wu⁹, Wei Hu¹⁰, Cheng Li⁶, Xiaole Fan¹¹, Liyun Jiang², Xiaofan Lu², Fangrong Yan²

Affiliations

¹ Department of Radiology, Zhujiang Hospital, Southern Medical University, Guangzhou, China.
² State Key Laboratory of Natural Medicines, Research Center of Biostatistics and Computational Pharmacy, China Pharmaceutical University, Nanjing, China.
³ Department of Radiology, The Affiliated Hospital of Nantong University, Nantong, China.
⁴ Department of Endocrinology, Tongren Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China.
⁵ Department of Neurology, Drum Tower Hospital, Medical School and The State Key Laboratory of Pharmaceutical Biotechnology, Institute of Brain Science, Nanjing University, Nanjing, China.
⁶ Research & Advanced Algorithm Department of HSW BU, Shanghai United Imaging Healthcare Co., Ltd., Shanghai, China.
⁷ Medical School of Nanjing University, Nanjing, China.
⁸ Calibration Physical Algorithm Department of CT BU, Shanghai United Imaging Healthcare Co., Ltd., Shanghai, China.
⁹ Clinical Workflow and Clinical Verification Department of CT BU, Shanghai United Imaging Healthcare Co., Ltd., Shanghai, China.
¹⁰ Department of CT BU, Shanghai United Imaging Healthcare Co., Ltd., Shanghai, China.
¹¹ Department of Radiology, The Second Affiliated Hospital of Nantong University, Nantong, China.

Abstract

Background: 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.

Methods: 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.

Results: 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 P < 0.001).

Conclusions: 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.

Keywords: accurate and repeatable images; automatic cranial scanning; computed tomography; deep learning; stroke.