Table 1_Machine learning-based prediction of ischemic cardio-cerebrovascular events after

1,954 patient records from the China Treatment Trial for Unruptured Intracranial Aneurysm (ChTUIA) registry, collected between December 2021 and December 2024. Yingchao He created this dataset to develop a machine learning framework for predicting ischemic cardio-cerebrovascular events within 6 months after endovascular or microsurgical treatment. The dataset includes 240 observed event cases (12.28%) and was used to evaluate nine predictive models.

Use Cases

• Training machine learning models for ischemic event prediction based on clinical registry data.
• Evaluating the triglyceride-glucose index as a metabolic marker for perioperative risk stratification.
• Benchmarking model performance across logistic regression and advanced algorithms like CatBoost.
• Conducting subgroup analyses on patients undergoing endovascular versus microsurgical treatment.

Strengths

• Dataset contains 1,954 patient records with a clear outcome measure of 240 ischemic events.
• Model development employed rigorous methods including feature selection and class imbalance handling.
• The optimal CatBoost model achieved a reported AUROC of 0.945, indicating strong predictive performance.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count for the specific analysis table is unknown, which may limit suitability assessment.
• Data reflects a specific geographic and temporal bias inherent to the single-source registry.

Provenance

Source

China Treatment Trial for Unruptured Intracranial Aneurysm (ChTUIA) registry.

Collection Method

Patients enrolled from a clinical registry who underwent microsurgical or endovascular treatment.

Time Range

December 2021 to December 2024, with a 6-month follow-up period.

Freshness

Last updated 2026-05-13 04:41:24; freshness should be verified.

Geography

Likely China, based on the registry name.