ChatPPG.
Join free

Paper

Individual-level Public-Private Decoupling for Continuous Blood Pressure Waveform Prediction (IPPD)

Under review2026Added Oct 5, 2026

  • AB

Authors: Anonymous (ICLR 2027 double-blind)

ICLR 2027 submission (under review)

Paper page (opens in a new tab) Cite

TL;DR

Splits PPG representations into statistically decorrelated Public (population) and Private (individual) features; predicts ABP as public waveform + individual residual. Dual-view contrastive learning on the Private branch separates within-person consistency from between-person variation. On VitalDB-360, without target-case calibration or fine-tuning: waveform RMSE 11.65 mmHg, MAE 9.91 mmHg across 30 held-out cases; claims best individual-level performance vs BP baselines.

  • Form factor: finger

Results on ChatPPG leaderboards · 1

All leaderboards
LeaderboardMethodResultRank
VitalDB RMSE (mmHg)Cuffless blood pressure estimationIndividual-level Public-Private Decoupling for C11.65RMSE (mmHg), lower is better#1 of 1

1 row is credited to this paper. Ranks apply within each leaderboard only, as reported by the source.

Tasks & signals

Tasks

Cuffless blood pressure estimationBlood pressureSignal denoising

Signals

PPGBP

Keywords

cuffless-BPPPG-to-ABPwaveform-reconstructioncalibration-freepersonalizationcontrastiveICLR-2027-sub

Datasets

How it works

How Method works: Setup → Method → TargetSetup: Signals (PPG, ABP); Form factor (finger). Method: Method (Contrastive learning · Masked reconstruction). Target: Blood pressure; Signal denoising.SetupSignals: PPG, ABPSignalsPPG, ABPForm factor: fingerForm factorfingerMethodMethodContrastive learning · Masked…TargetBlood pressureBlood pressureSignal denoisingSignal denoising

Cite this paper

APA

Individual-level Public-Private Decoupling for Continuous Blood Pressure Waveform Prediction (IPPD). (2026). ICLR 2027 submission. https://openreview.net/forum?id=a8h33qllxT

BibTeX

@misc{ippdpublicprivat2026individuallevel,
  title         = {{Individual-level Public-Private Decoupling for Continuous Blood Pressure Waveform Prediction (IPPD)}},
  year          = {2026},
  howpublished  = {ICLR 2027 submission},
  url           = {https://openreview.net/forum?id=a8h33qllxT}
}

Built from the hub record. Check it against the publisher before submitting.

Added to the ChatPPG research hub

Same tasks, data and authors

Papers

Models

Datasets