Toward human intervention-free clinical diagnosis of intracranial aneurysm via deep neural network
Zi-Hao Bo,
Hui Qiao,
Chong Tian,
Yuchen Guo,
Wuchao Li,
Tiantian Liang,
Dongxue Li,
Dan Liao,
Xianchun Zeng,
Leilei Mei,
Tianliang Shi,
Bo Wu,
Chao Huang,
Lu Liu,
Can Jin,
Qiping Guo,
Jun-Hai Yong,
Feng Xu,
Tijiang Zhang,
Rongpin Wang,
Qionghai Dai
Affiliations
Zi-Hao Bo
BNRist and School of Software, Tsinghua University, Beijing, Beijing 100084, China
Hui Qiao
BNRist and Department of Automation, Tsinghua University, Beijing, Beijing 100084, China; Institute of Brain and Cognitive Sciences, Tsinghua University, Beijing, Beijing 100084, China
Chong Tian
Department of Radiology and Guizhou Provincial Key Laboratory of Intelligent Medical Image Analysis and Precision Diagnosis, Guizhou Provincial People's Hospital, Guiyang, Guizhou 550002, China
Yuchen Guo
BNRist and Department of Automation, Tsinghua University, Beijing, Beijing 100084, China
Wuchao Li
Department of Radiology and Guizhou Provincial Key Laboratory of Intelligent Medical Image Analysis and Precision Diagnosis, Guizhou Provincial People's Hospital, Guiyang, Guizhou 550002, China
Tiantian Liang
Department of Radiology and Guizhou Provincial Key Laboratory of Intelligent Medical Image Analysis and Precision Diagnosis, Guizhou Provincial People's Hospital, Guiyang, Guizhou 550002, China
Dongxue Li
Department of Radiology and Guizhou Provincial Key Laboratory of Intelligent Medical Image Analysis and Precision Diagnosis, Guizhou Provincial People's Hospital, Guiyang, Guizhou 550002, China
Dan Liao
Department of Radiology and Guizhou Provincial Key Laboratory of Intelligent Medical Image Analysis and Precision Diagnosis, Guizhou Provincial People's Hospital, Guiyang, Guizhou 550002, China
Xianchun Zeng
Department of Radiology and Guizhou Provincial Key Laboratory of Intelligent Medical Image Analysis and Precision Diagnosis, Guizhou Provincial People's Hospital, Guiyang, Guizhou 550002, China
Leilei Mei
Department of Radiology, Affiliated Hospital of Zunyi Medical University, Zunyi, Guizhou 563000, China
Tianliang Shi
Department of Radiology, Tongren Municipal People's Hospital, Tongren, Guizhou 554300, China
Bo Wu
Department of Radiology, Tongren Municipal People's Hospital, Tongren, Guizhou 554300, China
Chao Huang
Department of Radiology, Tongren Municipal People's Hospital, Tongren, Guizhou 554300, China
Lu Liu
Department of Radiology, The Second People's Hospital of Guiyang, Guiyang, Guizhou 550002, China
Can Jin
Department of Radiology, The Second People's Hospital of Guiyang, Guiyang, Guizhou 550002, China
Qiping Guo
Department of Radiology, Xingyi Municipal People's Hospital, Xingyi, Guizhou 562400, China
Jun-Hai Yong
BNRist and School of Software, Tsinghua University, Beijing, Beijing 100084, China
Feng Xu
BNRist and School of Software, Tsinghua University, Beijing, Beijing 100084, China; Institute of Brain and Cognitive Sciences, Tsinghua University, Beijing, Beijing 100084, China; Corresponding author
Tijiang Zhang
Department of Radiology, Affiliated Hospital of Zunyi Medical University, Zunyi, Guizhou 563000, China; Corresponding author
Rongpin Wang
Department of Radiology and Guizhou Provincial Key Laboratory of Intelligent Medical Image Analysis and Precision Diagnosis, Guizhou Provincial People's Hospital, Guiyang, Guizhou 550002, China; Corresponding author
Qionghai Dai
BNRist and Department of Automation, Tsinghua University, Beijing, Beijing 100084, China; Institute of Brain and Cognitive Sciences, Tsinghua University, Beijing, Beijing 100084, China; Corresponding author
Summary: Intracranial aneurysm (IA) is an enormous threat to human health, which often results in nontraumatic subarachnoid hemorrhage or dismal prognosis. Diagnosing IAs on commonly used computed tomographic angiography (CTA) examinations remains laborious and time consuming, leading to error-prone results in clinical practice, especially for small targets. In this study, we propose a fully automatic deep-learning model for IA segmentation that can be applied to CTA images. Our model, called Global Localization-based IA Network (GLIA-Net), can incorporate the global localization prior and generates the fine-grain three-dimensional segmentation. GLIA-Net is trained and evaluated on a big internal dataset (1,338 scans from six institutions) and two external datasets. Evaluations show that our model exhibits good tolerance to different settings and achieves superior performance to other models. A clinical experiment further demonstrates the clinical utility of our technique, which helps radiologists in the diagnosis of IAs. The bigger picture: Intracranial aneurysms (IAs) are enormous threats to human health with a prevalence of approximately 4%. The rupture of IAs usually causes death or severe damage to the patients. To enhance the clinical diagnosis of IAs, we present a deep learning model (GLIA-Net) for IA detection and segmentation without laborious human intervention, which achieves superior diagnostic performance validated by quantitative evaluations as well as a sophisticated clinical study. We anticipate that the publicly released data and the artificial intelligence technique would help to transform the clinical diagnostics and precision treatments of cerebrovascular diseases. They may also revolutionize the landscape of healthcare and biomedical research in the future.