PaCX-MAE: Physiology-Augmented Chest X-Ray Masked Autoencoder
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913177928728576 |
|---|---|
| author | Liu, Yancheng Maeda, Kenichi Pancholy, Manan |
| author_facet | Liu, Yancheng Maeda, Kenichi Pancholy, Manan |
| contents | Clinical diagnosis often requires combining imaging with physiological measurements, yet deployed models typically operate on unimodal data. We present PaCX-MAE, a cross-modal distillation framework that injects physiological priors into chest X-ray (CXR) encoders while remaining strictly unimodal at inference. PaCX-MAE augments in-domain masked autoencoding with a dual contrastive-predictive objective, aligning CXR representations with paired ECG and laboratory embeddings. Extensive evaluation across nine benchmarks demonstrates consistent improvements over domain-specific MAE, particularly on physiology-dependent tasks (e.g., +2.7 AUROC on MedMod; +6.5 F1 on VinDr). The method proves highly label-efficient in the 1% regime and preserves anatomical fidelity, achieving parity with MAE on segmentation tasks. Zero-shot and attention analyses confirm that PaCX-MAE successfully learns to attend to physiological indicators, such as the cardiac silhouette, absent in standard visual pretraining. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2606_01537 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PaCX-MAE: Physiology-Augmented Chest X-Ray Masked Autoencoder Liu, Yancheng Maeda, Kenichi Pancholy, Manan Computer Vision and Pattern Recognition Machine Learning Clinical diagnosis often requires combining imaging with physiological measurements, yet deployed models typically operate on unimodal data. We present PaCX-MAE, a cross-modal distillation framework that injects physiological priors into chest X-ray (CXR) encoders while remaining strictly unimodal at inference. PaCX-MAE augments in-domain masked autoencoding with a dual contrastive-predictive objective, aligning CXR representations with paired ECG and laboratory embeddings. Extensive evaluation across nine benchmarks demonstrates consistent improvements over domain-specific MAE, particularly on physiology-dependent tasks (e.g., +2.7 AUROC on MedMod; +6.5 F1 on VinDr). The method proves highly label-efficient in the 1% regime and preserves anatomical fidelity, achieving parity with MAE on segmentation tasks. Zero-shot and attention analyses confirm that PaCX-MAE successfully learns to attend to physiological indicators, such as the cardiac silhouette, absent in standard visual pretraining. |
| title | PaCX-MAE: Physiology-Augmented Chest X-Ray Masked Autoencoder |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2606.01537 |