Prediction of Recurrence and Rupture Risk of Ruptured and Unruptured Intracranial Aneurysms of the Posterior...

Area of Science:

• Neurosurgery
• Radiology
• Data Science

Background:

• Posterior circulation intracranial aneurysms pose higher rupture risks and recurrence rates than anterior circulation types.
• Effective long-term management is challenged by these increased risks.
• Early detection of high-risk aneurysms is crucial for patient outcomes.

Purpose of the Study:

• To identify key risk factors for intracranial aneurysm recurrence and rupture.
• To define criteria for early detection of high-risk posterior circulation aneurysms.
• To utilize machine learning for predictive analysis.

Main Methods:

• Retrospective analysis of 229 patients with posterior circulation intracranial aneurysms (2008-2020).
• Application of machine learning algorithms for pattern detection and prediction.
• 10-fold cross-validation and Area Under the Curve (AUC) for model selection and performance evaluation.

Main Results:

• Machine learning model achieved an AUC of 0.74, with sensitivity 0.76, specificity 0.70, and accuracy 0.76.
• Significant predictors for recurrence: age, aneurysm size, arterial hypertension, nicotine consumption.
• Independent risk factors for early recurrence: multiple aneurysms, basilar artery localization.
• Predictive features for rupture: advanced age, basilar artery localization, atherosclerosis, irregular morphology, familial predisposition.

Conclusions:

• Machine learning effectively identified known and novel risk factors for posterior circulation aneurysm recurrence and rupture.
• Findings support the clinical utility of ML in risk stratification and early detection.
• Results provide a foundation for prospective studies and clinical application.