Tao Wengui, Li Shifu, Zeng Chudai, Chen Zhou, Huang Zheng, Chen Fenghua
Department of Neurosurgery, Xiangya Hospital, Central South University, 87 Xiangya Street, Changsha, Hunan 410008, China (W.T., S.L., C.Z., Z.C., Z.H., F.C.); National Clinical Research Center for Geriatric Disorders, Central South University, 87 Xiangya Street, Changsha, Hunan 410008, China (W.T., S.L., C.Z., Z.C., Z.H., F.C.); Hypothalamic-Pituitary Research Center, Xiangya Hospital, Central South University, 87 Xiangya Road, Changsha, Hunan 410008, China (W.T., S.L., C.Z., Z.C., Z.H., F.C.).
Department of Neurosurgery, Xiangya Hospital, Central South University, 87 Xiangya Street, Changsha, Hunan 410008, China (W.T., S.L., C.Z., Z.C., Z.H., F.C.); National Clinical Research Center for Geriatric Disorders, Central South University, 87 Xiangya Street, Changsha, Hunan 410008, China (W.T., S.L., C.Z., Z.C., Z.H., F.C.); Hypothalamic-Pituitary Research Center, Xiangya Hospital, Central South University, 87 Xiangya Road, Changsha, Hunan 410008, China (W.T., S.L., C.Z., Z.C., Z.H., F.C.).
Acad Radiol. 2024 Apr;31(4):1583-1593. doi: 10.1016/j.acra.2023.08.023. Epub 2023 Sep 30.
This study aims to develop the best diagnostic model for brain arteriovenous malformations (bAVMs) rupture by using machine learning (ML) algorithms.
We retrospectively included 353 adult patients with ruptured and unruptured bAVMs. The clinical and radiological data on patients were collected. The significant variables were selected using univariable logistic regression. We constructed and compared the predictive models using five supervised ML algorithms, multivariable logistic regression, and R2eDAVM scoring system. A complementary systematic review and meta-analysis of studies was aggregated to explore the potential predictors for bAVMs rupture.
We found that a small nidus size of <3 cm, deep and infratentorial location, longer filling time, and deep and single venous drainage were associated with a higher risk of bAVMs rupture. The multilayer perceptron model showed the best performance with an area under the curve value of 0.736 (95% CI 0.67-0.801) and 0.713 (95% CI 0.647-0.779) in the training and test dataset, respectively. In our pooled analysis, we also found that the male sex, a single feeding artery, hypertension, non-White race, low Spetzler-Martin grade, and coexisting aneurysms are risk factors for bAVMs rupture.
This study further demonstrated the clinical and angioarchitectural characteristics in predicting bAVMs hemorrhage.
本研究旨在通过使用机器学习（ML）算法开发用于脑动静脉畸形（bAVM）破裂的最佳诊断模型。
我们回顾性纳入了353例患有破裂和未破裂bAVM的成年患者。收集了患者的临床和放射学数据。使用单变量逻辑回归选择显著变量。我们使用五种监督式ML算法、多变量逻辑回归和R2eDAVM评分系统构建并比较了预测模型。汇总了一项关于研究的补充系统评价和荟萃分析，以探索bAVM破裂的潜在预测因素。
我们发现，病灶大小<3 cm、深部和幕下位置、较长的充盈时间以及深部和单一静脉引流与bAVM破裂风险较高相关。多层感知器模型在训练数据集和测试数据集中分别表现最佳，曲线下面积值分别为0.736（95%CI 0.67-0.801）和0.713（95%CI 0.647-0.779）。在我们的汇总分析中，我们还发现男性、单一供血动脉、高血压、非白种人、低Spetzler-Martin分级以及并存动脉瘤是bAVM破裂的危险因素。
本研究进一步证明了预测bAVM出血中的临床和血管构筑特征。
