aiqbench analysis
This research addresses the critical challenge of model generalization in medical AI by introducing a harmonized, interpretable feature set derived directly from raw ECG waveforms to mitigate vendor and protocol mismatches between datasets. By benchmarking XGBoost models on this unified representation against end-to-end deep learning approaches, the study demonstrates that consistent waveform-derived features can maintain robust cross-dataset performance while offering transparency for clinical deployment. The work provides a practical pathway toward reliable external validation of cardiovascular prediction tools across different hospital environments without relying solely on black-box neural networks.
Harmonized Interpretable ECG Waveform Features for Robust Cross-Dataset Clinical Prediction
arXiv:2607.23412v1 Announce Type: new Abstract: Electrocardiograms (ECGs) are widely used for cardiovascular risk prediction, yet models often fail to transfer across hospitals because of protocol, population, and measurement differences. We benchmark cross-dataset generalization on three tasks - heart failure classification, 30-day all-cause mortality, and 30-day mortality among sinus-rhythm ECGs - using two large cohorts (MIMIC-IV and the Alberta Cohort). To reduce vendor-specific measurement mismatch, we build a harmonized, interpretable feature representation computed directly from raw waveforms: FeatureDB morphology/heart-rate-variability summaries plus compact time-frequency descriptors (autoregressive and wavelet features). We train XGBoost models on this unified feature space and evaluate with patient-disjoint internal and bidirectional external testing. We pre-specify two hypotheses: (H1) external AUROC retains at least 90% of source-site internal AUROC under transfer, and (H2) internal AUROC of the harmonized feature set stays within 10% of dataset-native machine-measurement models. Across tasks, internal AUROC is 0.79-0.82 and cross-dataset AUROC is 0.74-0.78, with larger and direction-dependent AUPRC shifts under transfer. As an exploratory benchmark, an end-to-end ConvNeXt model trained directly on raw ECG waveforms with age and sex achieves higher internal AUROC, while the harmonized representation remains competitive in relative cross-dataset transfer stability. These findings show that a consistent waveform-derived feature interface preserves performance, supports realistic external validation, and provides a transparent alternative for cross-site clinical prediction.
Abstract:Electrocardiograms (ECGs) are widely used for cardiovascular risk prediction, yet models often fail to transfer across hospitals because of protocol, population, and measurement differences. We benchmark cross-dataset generalization on three tasks - heart failure classification, 30-day all-cause mortality, and 30-day mortality among sinus-rhythm ECGs - using two large cohorts (MIMIC-IV and the Alberta Cohort). To reduce vendor-specific measurement mismatch, we build a harmonized, interpretable feature representation computed directly from raw waveforms: FeatureDB morphology/heart-rate-variability summaries plus compact time-frequency descriptors (autoregressive and wavelet features). We train XGBoost models on this unified feature space and evaluate with patient-disjoint internal and bidirectional external testing. We pre-specify two hypotheses: (H1) external AUROC retains at least 90% of source-site internal AUROC under transfer, and (H2) internal AUROC of the harmonized feature set stays within 10% of dataset-native machine-measurement models. Across tasks, internal AUROC is 0.79-0.82 and cross-dataset AUROC is 0.74-0.78, with larger and direction-dependent AUPRC shifts under transfer. As an exploratory benchmark, an end-to-end ConvNeXt model trained directly on raw ECG waveforms with age and sex achieves higher internal AUROC, while the harmonized representation remains competitive in relative cross-dataset transfer stability. These findings show that a consistent waveform-derived feature interface preserves performance, supports realistic external validation, and provides a transparent alternative for cross-site clinical prediction.
Machine Learning (cs.LG)
arXiv:2607.23412 [cs.LG]
(or arXiv:2607.23412v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2607.23412 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 303 words
Related coverage
CAP: Towards PPG Universal Representation Learning with Patient-level Supervision
arXiv · Jun 16, 2026
Interpretable EEG biomarkers with bag-of-waves: Spatial and temporal waveform dictionaries for low-data regimes
arXiv · Jul 27, 2026
Multi-Task Learning for Non-Canonical Phoneme Recognition via Articulatory Feature Decomposition
arXiv · Aug 25, 2026
Multimodal LLMs Outperform Pathology Foundation Models in Cross-Domain Histological Similarity
arXiv · Sep 29, 2026