Clinically informed pseudo-colour encoding for QT/T-wave ECG explanation quality
Encode ECG beat images using a pseudo-colour representation that fills the region under the QT interval, where the colour progression is designed from clinical electrophysiology knowledge and mapped to a validated QT-nomogram risk model. Compare a monochrome input pipeline against the pseudo-colour pipeline while keeping the neural architecture otherwise identical, then evaluate explanation faithfulness, localisation, and stability across multiple explanation methods on an external cohort.
Discovered 2026-11-10 · Sources checked 2026-11-10
Why it matters
The approach treats interpretability as something that can be shaped by the model input representation, with the reported goal of making explanation maps more localised and stable in clinically relevant ECG regions (QT interval / T-wave morphology) rather than relying on hard-to-verify visual features.
What changed
The input representation changed: the monochrome heartbeat image was replaced with a pseudo-colour encoding that colours the area under the QT interval according to a clinically motivated, QT-nomogram-linked sequence. Model architecture and configuration were otherwise held constant, isolating the effect to the chromatic channel.
Where it helps
- Improving explanation localisation for ECG risk assessment models
- Designing clinically aligned input encodings for electrophysiology-driven tasks
- Testing representation changes to enhance stability of explanation methods
- Mapping time-interval regions (e.g., QT) to clinically validated risk scales
Evidence & reproduction
Reproduction not evaluated. A linked paper or repository does not establish independent reproduction.
Limitations & open questions
Pub-AI has not independently evaluated this method. Consult the linked sources for their evaluation setup and limitations.
