Atrial Fibrillation Recurrence Predictors with Cardiac CT Features

297 patient records from a study developing machine learning models to predict atrial fibrillation recurrence after catheter ablation. The dataset likely contains 37 clinical and cardiac computed tomography imaging features, including right atrial structural parameters. SHapley Additive exPlanations (SHAP) analysis was used to interpret the models, identifying key predictors like right atrial appendage short diameter and right atrial volume.

Use Cases

• Predict recurrence of paroxysmal atrial fibrillation based on right atrial appendage short diameter.
• Predict recurrence of persistent atrial fibrillation based on right atrial volume.
• Develop explainable clinical decision support tools for post-ablation prognosis.
• Analyze nonlinear relationships and interaction effects between clinical and imaging features.
• Compare performance of machine learning algorithms like XGBoost and LightGBM for medical prediction tasks.

Strengths

• Includes 297 patient records, split into 230 paroxysmal AF and 67 persistent AF cases.
• Models achieved high predictive performance with AUC scores of 0.831 for paroxysmal AF and 0.917 for persistent AF.
• Employed explainable AI techniques (SHAP) to interpret model decisions and identify top predictors.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data may reflect temporal or institutional bias inherent to the single retrospective study.

Provenance

Source

figshare

Collection Method

Retrospective collection of clinical and cardiac CT imaging features from patients undergoing radiofrequency catheter ablation.

Freshness

Last updated 2026-04-29 05:57:08; freshness should be verified.