Table 1_Development of a nomogram for predicting wall irregular pulsation and risk stratification in intracranial aneurysms.docx

Model performance was evaluated using receiver operating characteristic analysis.</p>Results<p>We included 202 patients with 262 unruptured aneurysms. Multivariable analysis revealed four independent predictors of the presence of wall irregular pulsation: aneurysm location at ACA/PCom/Posterior (OR: 2.17, 95% CI 1.21–3.88; p = 0.0099), presence of daughter sac (OR: 5.15, 95% CI 1.91–13.86; p = 0.001), aspect ratio (OR: 1.80, 95% CI 1.09–2.99; p = 0.022), and neck size (OR: 1.23, 95% CI 1.02–1.48; p = 0.030).

The developed model demonstrated good discrimination, achieving a sensitivity of 88.5%, specificity of 71.2%, overall accuracy of 76.9%, and an area under the curve (AUC) of 0.808 (95% CI, 0.704–0.913). Further validation using an independent cohort yielded comparable performance, with a sensitivity of 76.9%, specificity of 88.9%, overall accuracy of 81.8%, and an AUC of 0.855 (95% CI, 0.694–1.000). A nomogram was constructed to visualize the model.</p>Conclusion<p>This study presents a logistic regression-based model for predicting aneurysm wall irregular pulsation.

The nomogram visually assesses individualized risks, demonstrating potential for future clinical implementation in bedside decision-making.</p>