How the risk estimate is built
The platform uses established cardiovascular predictors inspired by Framingham and Pooled Cohort frameworks, then translates them into an interpretable risk score for decision support.
How the formula works
Age contributes baseline vascular risk and calibration by decade.
ROC-AUC: 0.79 | Accuracy: 72% | Trained on 70,000 records from the public Kaggle cardiovascular disease dataset (49,000 train / 21,000 held-out test). See the model card for the full breakdown.
This is a lifestyle risk-screening model, not a diagnostic tool -- it uses self-reportable factors (age, sex, BMI, blood pressure, cholesterol/glucose category, smoking, alcohol, activity) and does not include ECG, imaging, lab-confirmed lipid panels, or family history.
Every prediction is accompanied by real SHAP (SHapley Additive exPlanations) values showing which factors pushed your risk estimate up or down, not just a static list of "important" features.
References (APA)
Ulianova, S. (2019). Cardiovascular Disease dataset [Data set]. Kaggle. https://www.kaggle.com/datasets/sulianova/cardiovascular-disease-dataset
Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 30.
World Health Organization. (2025). Cardiovascular diseases fact sheet.