Clinical Research Directory

Prospective Observational Cohort Study for Predicting Unruptured Intracranial Aneurysm Rupture Risk Using Blood Pressure Variability and 3D Structural Evolution

This is a single-center, prospective, longitudinal observational cohort study in adults with unruptured intracranial aneurysms (UIAs) who are managed by conservative observation rather than immediate surgery or endovascular treatment. The study aims to build a time-series dataset combining blood pressure variability (BPV) and three-dimensional (3D) aneurysm morphological evolution, and to develop deep-learning models using long short-term memory (LSTM) and Transformer networks to predict aneurysm rupture risk. A secondary aim is to compare the new models with traditional scores such as PHASES and to explore how BPV relates to 3D aneurysm change.Participants aged 18-80 years with UIA confirmed by CTA, MRA, or DSA and judged suitable for conservative follow-up will be enrolled consecutively at the study site. Those with imaging contraindications, other major cerebrovascular diseases such as moyamoya or arteriovenous malformation, or severe organ dysfunction are excluded. No study-driven intervention is imposed; routine clinical care continues. Baseline data include demographics, hypertension history, laboratory tests, 24-hour ambulatory blood pressure, home blood pressure for 7 days, optional wearable monitoring, and CTA/MRA/DSA for 3D aneurysm parameters such as volume, surface area, compactness, and sphericity. BPV is summarized using standard deviation, coefficient of variation, and variability independent of the mean.Participants scheduled for in-hospital visits provide written informed consent and are followed at 6, 12, 24, 36, 48, 72, 96, and 120 months with repeated BPV measurement and imaging. Participants who already have at least two qualifying imaging and blood-pressure records before enrollment and will not return for in-person visits provide verbal consent and are followed by telephone every 6-12 months to confirm health status and rupture/growth events. All data are entered into a REDCap system with automated and manual quality checks. A historical institutional dataset (CAIASA) may be used only for algorithm pretraining and is not counted as the prospective enrollment target.The primary outcome is UIA rupture confirmed by CT or surgery. Secondary outcomes include aneurysm growth defined as diameter increase ≥1 mm, changes in 3D morphological parameters over time, BPV trends, and incremental predictive performance versus PHASES using AUC, IDI, and NRI with bootstrapping validation. A data safety monitoring committee oversees safety and data integrity. Findings are expected to support individualized follow-up and treatment decisions for people with unruptured aneurysms.