O27 Radiomic-based predictive nomogram for assessing the risk of intracranial aneurysms

<h3>Introduction</h3> Aneurysm wall enhancement (AWE) is a potential biomarker for risk stratification of intracranial aneurysms (IAs)¹. Radiomics provides a refined approach to quantify and characterize AWE’s textural features². AWE and known risk factors for IAs rupture can improve the accuracy of a nomogram in detecting symptomatic aneurysms. <h3>Aim of Study</h3> Offer a nomogram using AWE, Radiomic and known risk factor in detecting symptomatic aneurysms. <h3>Methods</h3> Ninety patients with 104 IAs (29 symptomatic and 75 asymptomatic) underwent 3T-High-resolution MRI to quantify AWE. The assessment of AWE was performed with 3D-AWE mapping and composite radiomics-based score (RadScore). The dataset was split into training and testing subsets. The testing set was used to build two different nomograms using each modality of AWE assessment combined with patient’s demographic information and aneurysm morphological data. Finally, each nomogram was evaluated on an independent testing set. <h3>Results</h3> 87 radiomic features significantly distinguished symptomatic from asymptomatic IAs. The initial logistic regression model demonstrated an accuracy of 63%, with a sensitivity of 67%, specificity of 62%, and an area under the curve (AUC) of 0.76. The RadScore nomogram improved accuracy to 77.1%, achieving a sensitivity of 89%, specificity of 73%, and an AUC of 0.83. The 3D-AWE Mapping nomogram achieved an accuracy of 63%, sensitivity of 78%, specificity of 58%, and AUCs of 0.771. <h3>Conclusion</h3> Combining radiomic analysis with patient demographic data markedly improves the accuracy in detecting symptomatic IAs. Nomograms that include patient’s clinical risk factors, aneurysms’ morphological features and AWE data, offer a robust approach for assessing the symptomatic status of IAs.