Prediction of cerebral aneurysm rupture using a point cloud neural network

• Published in: Journal of neurointerventional surgery Vol. 15; no. 4; pp. 380 - 386
• Main Authors: Luo, Xiaoyuan, Wang, Jienan, Liang, Xinmei, Yan, Lei, Chen, XinHua, He, Jian, Luo, Jing, Zhao, Bing, He, Guangchen, Wang, Manning, Zhu, Yueqi
• Format: Journal Article
• Language: English
• Published: BMA House, Tavistock Square, London, WC1H 9JR BMJ Publishing Group Ltd 01.04.2023; BMJ Publishing Group LTD
• Subjects: Accuracy; Aneurysm; Aneurysms; Angiography; Artificial intelligence; CT Angiography; Digital imaging; Humans; Intervention; Intracranial Aneurysm - diagnostic imaging; Magnetic Resonance Angiography; Medical imaging; Morphology; Neural networks; Neural Networks, Computer; Neuroimaging; Prospective Studies; Radiomics; Retrospective Studies; Intervention; Magnetic Resonance Angiography; CT Angiography; Aneurysm
• ISSN: 1759-8478; 1759-8486
• DOI: 10.1136/neurintsurg-2022-018655

ObjectiveAccurate prediction of cerebral aneurysm (CA) rupture is of great significance. We intended to evaluate the accuracy of the point cloud neural network (PC-NN) in predicting CA rupture using MR angiography (MRA) and CT angiography (CTA) data.Methods418 CAs in 411 consecutive patients confirmed by CTA (n=180) or MRA (n=238) in a single hospital were retrospectively analyzed. A PC-NN aneurysm model with/without parent artery involvement was used for CA rupture prediction and compared with ridge regression, support vector machine (SVM) and neural network (NN) models based on radiomics features. Furthermore, the performance of the trained PC-NN and radiomics-based models was prospectively evaluated in 258 CAs of 254 patients from five external centers.ResultsIn the internal test data, the area under the curve (AUC) of the PC-NN model trained with parent artery (AUC=0.913) was significantly higher than that of the PC-NN model trained without parent artery (AUC=0.851; p=0.041) and of the ridge regression (AUC=0.803; p=0.019), SVM (AUC=0.788; p=0.013) and NN (AUC=0.805; p=0.023) radiomics-based models. Additionally, the PC-NN model trained with MRA source data achieved a higher prediction accuracy (AUC=0.936) than that trained with CTA source data (AUC=0.824; p=0.043). In external data of prospective cohort patients, the AUC of PC-NN was 0.835, significantly higher than ridge regression (0.692; p<0.001), SVM (0.701; p<0.001) and NN (0.681; p<0.001) models.ConclusionPC-NNs can achieve more accurate CA rupture prediction than traditional radiomics-based models. Furthermore, the performance of the PC-NN model trained with MRA data was superior to that trained with CTA data.