基于磁敏感加权 MR 序列的深度学习用于检测脑微出血和分类脑小血管病。
Deep learning based on susceptibility-weighted MR sequence for detecting cerebral microbleeds and classifying cerebral small vessel disease.
机构信息
Department of Neurology, The Third Affiliated Hospital of Sun Yat-Sen University, Sun Yat-Sen University, No. 600 Tianhe Road, Guangzhou, 510630, People's Republic of China.
Center for Artificial Intelligence in Medicine, Research Institute of Tsinghua, Pearl River Delta, Tsinghua University, No. 98 Xiangxue 8Th Road, Guangzhou, 510700, People's Republic of China.
出版信息
Biomed Eng Online. 2023 Oct 17;22(1):99. doi: 10.1186/s12938-023-01164-1.
BACKGROUND
Cerebral microbleeds (CMBs) serve as neuroimaging biomarkers to assess risk of intracerebral hemorrhage and diagnose cerebral small vessel disease (CSVD). Therefore, detecting CMBs can evaluate the risk of intracerebral hemorrhage and use its presence to support CSVD classification, both are conducive to optimizing CSVD management. This study aimed to develop and test a deep learning (DL) model based on susceptibility-weighted MR sequence (SWS) to detect CMBs and classify CSVD to assist neurologists in optimizing CSVD management. Patients with arteriolosclerosis (aSVD), cerebral amyloid angiopathy (CAA), and cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) treated at three centers were enrolled between January 2017 and May 2022 in this retrospective study. The SWSs of patients from two centers were used as the development set, and the SWSs of patients from the remaining center were used as the external test set. The DL model contains a Mask R-CNN for detecting CMBs and a multi-instance learning (MIL) network for classifying CSVD. The metrics for model performance included intersection over union (IoU), Dice score, recall, confusion matrices, receiver operating characteristic curve (ROC) analysis, accuracy, precision, and F1-score.
RESULTS
A total of 364 SWS were recruited, including 336 in the development set and 28 in the external test set. IoU for the model was 0.523 ± 0.319, Dice score 0.627 ± 0.296, and recall 0.706 ± 0.365 for CMBs detection in the external test set. For CSVD classification, the model achieved a weighted-average AUC of 0.908 (95% CI 0.895-0.921), accuracy of 0.819 (95% CI 0.768-0.870), weighted-average precision of 0.864 (95% CI 0.831-0.897), and weighted-average F1-score of 0.829 (95% CI 0.782-0.876) in the external set, outperforming the performance of the neurologist group.
CONCLUSION
The DL model based on SWS can detect CMBs and classify CSVD, thereby assisting neurologists in optimizing CSVD management.
背景
脑微出血 (CMB) 作为神经影像学生物标志物，可用于评估脑出血风险和诊断脑小血管疾病 (CSVD)。因此，检测 CMB 可以评估脑出血风险，并将其存在用于支持 CSVD 分类，这两者都有助于优化 CSVD 管理。本研究旨在开发和测试一种基于磁敏感加权磁共振序列 (SWS) 的深度学习 (DL) 模型，以检测 CMB 并对 CSVD 进行分类，以帮助神经科医生优化 CSVD 管理。本回顾性研究纳入了 2017 年 1 月至 2022 年 5 月期间在三个中心治疗的动脉粥样硬化性 (aSVD)、脑淀粉样血管病 (CAA) 和伴有皮质下梗死和白质脑病的常染色体显性脑动脉病 (CADASIL) 患者。来自两个中心的患者的 SWS 用于开发集，来自其余中心的患者的 SWS 用于外部测试集。DL 模型包含一个用于检测 CMB 的 Mask R-CNN 和一个用于分类 CSVD 的多实例学习 (MIL) 网络。模型性能的指标包括交并比 (IoU)、Dice 评分、召回率、混淆矩阵、受试者工作特征曲线 (ROC) 分析、准确性、精度和 F1 评分。
结果
共纳入 364 例 SWS，其中 336 例来自开发集，28 例来自外部测试集。外部测试集的模型的 IoU 为 0.523 ± 0.319，Dice 评分为 0.627 ± 0.296，CMB 检测的召回率为 0.706 ± 0.365。对于 CSVD 分类，模型在外部集的加权平均 AUC 为 0.908(95%CI 0.895-0.921)，准确性为 0.819(95%CI 0.768-0.870)，加权平均精度为 0.864(95%CI 0.831-0.897)，加权平均 F1 得分为 0.829(95%CI 0.782-0.876)，均优于神经科医生组的表现。
结论
基于 SWS 的 DL 模型可以检测 CMB 并对 CSVD 进行分类，从而帮助神经科医生优化 CSVD 管理。
Zotero 插件