Normal-Anchored SPD Transport for Source-Free ECG Adaptation

A new geometric transport method adapts ECG beat classifiers to new patients without labeled target data, addressing the extreme class imbalance of real Holter and patch recordings.

3 November 2026 · 1 min read · 1 source

ECG beat classifiers often degrade when applied to new patients or devices. Source-free adaptation methods try to adjust the model using unlabeled target data, but many assume that the target batch is class-balanced.

Real Holter and patch ECG recordings break that assumption: the label-free prefix is 87.9–98.6% normal beats. The preprint introduces normal-anchored SPD geometric transport, a source-free adaptation approach designed for this extreme imbalance.

The paper is described only in its arXiv abstract, so no comparison results are available from this source. It frames the method as a practical fix for a gap in current adaptation techniques.

Source

01Transport What Shares the Conduction Pathway: Normal-Anchored SPD Geometric Transport for Source-Free Patient Adaptation of Holter and Patch ECGarXiv

Normal-Anchored SPD Transport for Source-Free ECG Adaptation

Source

More in Research Digest

Polynomial-Time Learning Shown for Positive ReLU Networks Under Gaussian Inputs

Sequence-Level Policy Optimization with Conformal Prediction for Reasoning LLMs

ClinSDT adds LLM semantic guidance to Decision Transformers for sepsis

Optimal Sample Size Found for Pointwise Replicable Learning