PaperarXiv:2611.11871
Integrating Deep PPG Representations and Digital Biomarkers for Aortic Stenosis Detection with Peri-TAVI Assessment
arXiv preprint (eess.SP)
TL;DR
PulseAS screens for high-grade aortic stenosis from 15 s of fingertip PPG by combining frozen AnyPPG embeddings, pyPPG morphology and clinical data. Patient-level CV AUROC 0.818 (Hong Kong EDU, n=1,209) and 0.838 (MIMIC-III, n=2,970); MIMIC to EDU transfer 0.716, 0.813 after local adaptation. MIT code.
- Form factor: clinical (fingertip PPG, CONTEC CMS50E at 60 Hz; MIMIC-III bedside PLETH)
Abstract
Access to echocardiography is often limited, delaying recognition of progressive aortic stenosis (AS). A brief, low-cost physiological measurement obtained during routine care could help identify patients who may warrant earlier imaging. Here, we present PulseAS, integrating deep photoplethysmography (PPG) representations, waveform biomarkers and clinical information for high-grade AS screening. Three complementary analyses assessed (1) validation and transfer across cohorts differing in geography and ethnic composition, (2) score-based enrichment of AS and pre-transcatheter aortic valve implantation (TAVI) cases, and (3) preoperative and postoperative pulse changes. Using 15-s recordings, separate patient-level cross-validation in EDU (n = 1,209) and MIMIC-III (n = 2,970) yielded AUROCs of 0.818 (PulseAS-XGB) and 0.838 (PulseAS-L2), respectively. In external validation from MIMIC-III to EDU, PulseAS-XGB achieved an AUROC of 0.716, improving to 0.813 after local adaptation. The highest-scoring 10% captured 45.2% of all high-grade AS cases in EDU and 41.3% in MIMIC-III, and 56.9% of pre-TAVI cases in EDU. Notably, in the EDU TAVI subgroup, PPG waveforms showed earlier systolic peaks and narrower systolic contours following intervention. These findings support clinical-PPG risk stratification in new populations and warrant prospective evaluation of echocardiographic prioritisation.
Verbatim from arXiv (opens in a new tab) · arXiv metadata (abstracts are released under CC0)
Results on ChatPPG leaderboards
All leaderboardsNo results from this paper on ChatPPG leaderboards yet. Browse the leaderboards.
Tasks & signals
Signals
Keywords
How it works
Cite this paper
Wang, H., Ding, C., Xie, D., Yu, H., Nie, G., Hong, S., Yan, B. P., & So, K. C.-Y. (2026). Integrating Deep PPG Representations and Digital Biomarkers for Aortic Stenosis Detection with Peri-TAVI Assessment. arXiv preprint. https://arxiv.org/abs/2611.11871
Wang, Haowan, Cheng Ding, Donglin Xie, Huizi Yu, Guangkun Nie, Shenda Hong, Bryan P Yan, and Kent Chak-yu So. 2026. "Integrating Deep PPG Representations and Digital Biomarkers for Aortic Stenosis Detection with Peri-TAVI Assessment." arXiv preprint. https://arxiv.org/abs/2611.11871.
Wang, H., Ding, C., Xie, D., Yu, H., Nie, G., Hong, S., Yan, B. P. and So, K. C.-Y. (2026) 'Integrating Deep PPG Representations and Digital Biomarkers for Aortic Stenosis Detection with Peri-TAVI Assessment', arXiv preprint. Available at: https://arxiv.org/abs/2611.11871
Wang, Haowan, et al. "Integrating Deep PPG Representations and Digital Biomarkers for Aortic Stenosis Detection with Peri-TAVI Assessment." arXiv preprint, 2026. https://arxiv.org/abs/2611.11871.
H. Wang et al., "Integrating Deep PPG Representations and Digital Biomarkers for Aortic Stenosis Detection with Peri-TAVI Assessment," arXiv preprint, 2026, [Online]. Available: https://arxiv.org/abs/2611.11871.
@misc{wang2026integrating,
title = {{Integrating Deep PPG Representations and Digital Biomarkers for Aortic Stenosis Detection with Peri-TAVI Assessment}},
author = {Wang, Haowan and Ding, Cheng and Xie, Donglin and Yu, Huizi and Nie, Guangkun and Hong, Shenda and Yan, Bryan P. and So, Kent Chak-yu},
year = {2026},
howpublished = {arXiv preprint},
eprint = {2611.11871},
archivePrefix = {arXiv}
}Added to the ChatPPG research hub
