New AI model protects patient privacy in ECG data

▤ Summary

Researchers at the University of Kansas developed a privacy-preserving AI model for ECG data that reduces exposure of sensitive attributes such as age, sex, race, and identity. The model still predicts clinically useful outcomes like left ventricular ejection fraction, heart disease risk, and five-year mortality. The work addresses a real healthcare data-sharing problem by making ECG analytics safer for hospitals and research institutions. The team says it could support broader collaboration and future AI development without compromising patient privacy.

▦ Linked Companies