Division project · Active since 2024
Generative and predictive models over the electrocardiogram: reconstructing the full twelve-lead ECG from a minimal set of leads, and stratifying acute coronary syndrome risk from the admission ECG.
Overview
The electrocardiogram is the most widely collected cardiac signal and the least exploited. Two strands develop deep learning over it. The first asks whether a small number of leads, of the kind a wearable or a single-lead device can record, carries enough information to reconstruct the full twelve-lead signal. The second uses admission ECG features together with vital signs to identify acute coronary syndrome patients at high risk of in-hospital mortality and adverse events.
- Status
- Active since 2024
- Program
- Division project
- Team
- 1 member
Lines of work
2 strands
01
Minimal-lead ECG reconstruction
This project develops generative deep learning models to reconstruct multi-lead ECG signals from minimal input leads, assessing whether a small number of key ECG recordings can capture sufficient signal and diagnostic information for robust clinical interpretation and scalable cardiac monitoring.
No publication yet.
02
Risk stratification for acute coronary syndrome
Machine learning models identify high-risk acute coronary syndrome patients using admission ECG features and vital signs to predict in-hospital mortality and clinical deterioration, with interpretable outputs that guide triage and escalation decisions in acute cardiovascular care.
No publication yet.
People
