Prediction of bleb formation in intracranial aneurysms using machine learning models based on aneurysm hemodynamics, geometry, location, and patient population.
Prediction of bleb formation in intracranial aneurysms using machine learning models based on aneurysm hemodynamics, geometry, location, and patient population.
复制标题
DOI：
10.1136/neurintsurg-2021-017976
复制
发表时间：
2022-10
影响因子：
4.8
通讯作者：
Cebral, Juan R.
中科院分区：
文献类型：
--
作者：
Salimi Ashkezari, Seyedeh Fatemeh;Mut, Fernando;Slawski, Martin;Cheng, Boyle;Yu, Alexander K.;White, Tim G.;Woo, Henry H.;Koch, Matthew J.;Amin-Hanjani, Sepideh;Charbel, Fady T.;Rezai Jahromi, Behnam;Niemela, Mika;Koivisto, Timo;Frosen, Juhana;Tobe, Yasutaka;Maiti, Spandan;Robertson, Anne M.;Cebral, Juan R.
颅内动脉瘤（IA）中存在滤过泡是不稳定性和脆弱性的已知指征。我们的目标是根据血流动力学、几何形状、解剖位置和患者人群开发和评价颅内动脉瘤中滤过泡发展的预测模型。2395例IA的横截面数据（一个时间点）用于使用机器学习（随机森林、支持向量机、逻辑回归、k-最近邻和装袋）训练水泡形成模型。使用基于图像的计算流体动力学表征动脉瘤血流动力学和几何形状。使用包含266个动脉瘤的单独数据集进行模型评价。通过接收操作特征下的面积（AUC）、真阳性率（TPR）、假阳性率（FPR）、精密度和平衡准确度来量化模型性能。最终模型保留了18个变量，包括血流动力学、几何形状、位置、多重性和形态学参数以及患者人群。一般来说，强烈和集中的流入射流，高速，复杂和不稳定的流动模式，以及集中，振荡和不均匀的壁面剪应力模式，如沿着更大，更长，更扭曲的形状与气泡的形成。验证集上的最佳性能由随机森林模型实现（AUC=0.82，TPR= 91%，FPR= 36%，误分类误差=27%）。基于泡形成之前的动脉瘤特征类似于从血管重建中获得的那些特征，其泡几乎被去除，ML模型能够以良好的准确性识别易于形成泡的动脉瘤。在等待进一步验证与纵向数据，这些模型可能被证明是有价值的评估的倾向，进展到脆弱的国家和潜在的破裂。
Bleb presence in intracranial aneurysms (IAs) is a known indication of instability and vulnerability. Our objective was to develop and evaluate predictive models of bleb development in IAs based on hemodynamics, geometry, anatomical location, and patient population. Cross-sectional data (one time-point) of 2395 IAs was used for training bleb formation models using machine learning (random forest, support vector machine, logistic regression, k-nearest neighbor, and bagging). Aneurysm hemodynamics and geometry were characterized using image-based computational fluid dynamics. A separate dataset with 266 aneurysms was used for model evaluation. Model performance was quantified by the area under the receiving operating characteristic (AUC), true positive rate (TPR), false positive rate (FPR), precision, and balanced accuracy. The final model retained 18 variables including hemodynamic, geometrical, location, multiplicity, and morphology parameters, and patient population. Generally, strong and concentrated inflow jets, high speed, complex and unstable flow patterns, and concentrated, oscillatory, and heterogeneous wall shear stress patterns as along with larger, more elongated, and more distorted shapes were associated with bleb formation. The best performance on the validation set was achieved by the random forest model (AUC=0.82, TPR=91%, FPR=36%, misclassification error=27%). Based on the premise that aneurysm characteristics prior to bleb formation resemble those derived from vascular reconstructions with their blebs virtually removed, ML models are capable of identifying aneurysms prone to bleb development with good accuracy. Pending further validation with longitudinal data, these models may prove valuable for assessing the IAs propensity of progressing to vulnerable states and potentially rupturing.
登录
查看更多内容
影响因子：
4.8
作者：
Salimi Ashkezari SF;Detmer FJ;Mut F;Chung BJ;Yu AK;Stapleton CJ;See AP;Amin-Hanjani S;Charbel FT;Rezai Jahromi B;Niemelä M;Frösen J;Zhou J;Maiti S;Robertson AM;Cebral JR
通讯作者：
Cebral JR
影响因子：
4.6
作者：
Heo, Jaehyuk;Park, Sang Jun;Kim, Tackeun
通讯作者：
Kim, Tackeun
影响因子：
8.3
作者：
Lindgren, Antti E.;Koivisto, Timo;Frosen, Juhana
通讯作者：
Frosen, Juhana
影响因子：
10.6
作者：
Cebral, JR;Castro, MA;Frangi, AF
通讯作者：
Frangi, AF
影响因子：
4.4
作者：
Nakao, Takahiro;Hanaoka, Shouhei;Abe, Osamu
通讯作者：
Abe, Osamu