Predicting intracranial aneurysm rupture risk and intervention outcomes using denoising-enhanced CT Angiography.
Authors
Affiliations (1)
Affiliations (1)
- Department of Neurosurgery, General Hospital of Nothern Theater Command, Shenyang 110016, China.
Abstract
To develop a comprehensive multi-modal framework for assessing the rupture risk of intracranial aneurysms and predicting intervention outcomes. In addition, it seeks to a novel denoising algorithm to enhance the quality of CTA images, thereby improving morphological profiling. This retrospective multicentre study included 352 patients who underwent CTA with intracranial aneurysm. A multi-modal framework was developed, integrating three complementary feature sets: clinical variables, radiomic texture features and deep learning-derived aneurysm morphological data. A novel denoising algorithm was applied to improve image quality, thereby enhancing prediction performance. Model validation was performed through cross-validation, utilizing multiple endpoints, including the Hunt-Hess, WFNS and Modified Rankin Scale grading systems. The multi-modal framework demonstrated robust performance, achieving an AUC of 0.896 [0.819-0.973] for aneurysm rupture prediction, outperforming conventional single-modality models (radiomics-based model: 0.752 [0.693-0.809]; deep learningbased model: 0.823 [0.789-0.827]). Incorporating the denoising technique further enhanced performance, with the AUC for rupture prediction rising to 0.908 [0.836-0.981]. In clinical grading tasks, the model showed strong efficacy, achieving an AUC of 0.907 [0.845-0.968] for Hunt-Hess grading, 0.883 [0.662-0.988] for WFNS grading and 0.926 [0.879-0.973] for Modified Rankin Scale. Our system demonstrates promising performance in predicting aneurysm rupture and clinical grading assessments, indicating its potential for comprehensive aneurysm evaluation. Moreover, the proposed denoising method effectively mitigates noise interference, enhances morphological edge features, and improves the accuracy of existing models. AUC＝ Area under the curve; DSA＝ Digital subtraction angiography; IAs＝ Intracranial aneurysms; ICC＝ Intraclass correlation coefficient; ROC＝ Receiver operating characteristic; ROI＝ Region of interest; SAH＝ Subarachnoid hemorrhage; WFNS＝ World Federation of neurosurgical societies.