Diagnosis of intracranial aneurysms by computed tomography angiography using deep learning-based detection and segmentation

作者：Wei You, Junqiang Feng, Jing Lu, Ting Chen, Xinke Liu, Zhenhua Wu, Guoyang Gong, Yutong Sui, Yanwen Wang, Yifan Zhang, Wanxing Ye, Xiheng Chen, Jian Lv, Dachao Wei, Yudi Tang, Dingwei Deng, Siming Gui, Lin Jun, Peike Chen, Ziyao Wang, Wentao Gong, Yang Wang, Chengcheng Zhu, Yue Zhang, David Saloner, Dimitrios Mitsouras, Sheng Guan, Youxiang Li, Youxiang Li, Yuhua Jiang, Yan Wang · 发表于：Journal of NeuroInterventional Surgery · 年份：2024 · DOI：10.1136/jnis-2023-021022 · 被引用次数：13 · 研究领域：Intracranial Aneurysms: Treatment and Complications、Retinal Imaging and Analysis、Cerebrovascular and Carotid Artery Diseases

BACKGROUND: Detecting and segmenting intracranial aneurysms (IAs) from angiographic images is a laborious task. OBJECTIVE: To evaluates a novel deep-learning algorithm, named vessel attention (VA)-Unet, for the efficient detection and segmentation of IAs. METHODS: This retrospective study was conducted using head CT angiography (CTA) examinations depicting IAs from two hospitals in China between 2010 and 2021. Training included cases with subarachnoid hemorrhage (SAH) and arterial stenosis, common accompanying vascular abnormalities. Testing was performed in cohorts with reference-standard digital subtraction angiography (cohort 1), with SAH (cohort 2), acquired outside the time interval of training data (cohort 3), and an external dataset (cohort 4). The algorithm's performance was evaluated using sensitivity, recall, false positives per case (FPs/case), and Dice coefficient, with manual segmentation as the reference standard. RESULTS: The study included 3190 CTA scans with 4124 IAs. Sensitivity, recall, and FPs/case for detection of IAs were, respectively, 98.58%, 96.17%, and 2.08 in cohort 1; 95.00%, 88.8%, and 3.62 in cohort 2; 96.00%, 93.77%, and 2.60 in cohort 3; and, 96.17%, 94.05%, and 3.60 in external cohort 4. The segmentation accuracy, as measured by the Dice coefficient, was 0.78, 0.71, 0.71, and 0.66 for cohorts 1-4, respectively. VA-Unet detection recall and FPs/case and segmentation accuracy were affected by several clinical factors, including aneurysm size, bi...