Aortic stenosis cohort, 200-300 confirmed cases with paired ECG and echo

Overview

200 to 300 confirmed moderate-to-severe aortic stenosis cases, for external validation of an ECG-based AS detection model. Per case: a 12-lead ECG paired with an echocardiogram acquired within 6 months of it. Echo must carry the numeric measurements: aortic valve area, mean gradient, peak velocity and dimensionless index. Where these are missing or inconsistent, raw Doppler images must be included so the values can be remeasured. Labels used as ground truth must be adjudicated by blinded expert readers with predefined qualifications, a documented disagreement-resolution process and a full audit trail from label back to source. Also required alongside the cases: the ECG manufacturer breakdown across the cohort, the count of echo-ECG pairs within six months, and the confirmed AS case count. Multi-site sourcing and demographic diversity are an advantage, since the model is being validated beyond a single centre.

Progress

Data Specifications

Use Cases

• •Training and validating Medical imaging AI/ML models
• •Benchmarking Medical imaging detection and segmentation algorithms
• •Building de-identified Medical imaging research datasets for academic studies
• •Augmenting existing Medical imaging datasets to reduce class imbalance