ChatPPG.
Join free

Paperdoi:10.5281/zenodo.23513090

Where Do Prediction Intervals Fail for Calibration-Free Cuffless Blood Pressure? A Patient-Level Analysis Within and Across Datasets

Venue listed2026Added Oct 21, 2026

Submitted to Physiological Measurement (per README); analysis code v1.1.0 on Zenodo (concept DOI 10.5281/zenodo.23513090, version DOI 10.5281/zenodo.23527711, 2026-10-21)

Publisher (opens in a new tab) Cite

Open codeLicence: Code MIT (LICENSE in repo; Zenodo mit-license); data_lists/ follow source-database licences

TL;DR

No summary in the hub record yet.

  • Form factor: clinical_finger
Caveats
  • Nominal-90 % split-conformal coverage 26.5 % at SBP ≥ 160 and 68.9 % on MIMIC-III external; DBP not transcribed.

Abstract

Analysis code for a study of conformal prediction intervals for blood pressure estimated from the photoplethysmogram (PPG) alone, without per-person calibration. It evaluates patient-level coverage within VitalDB and on MIMIC-III. Data are not included; the code downloads them from the original sources.

Verbatim from DataCite (opens in a new tab) · Licence: https://opensource.org/licenses/MIT

Results on ChatPPG leaderboards · 1

All leaderboards
LeaderboardMethodResultRank
VitalDB MAE (mmHg)Cuffless blood pressure estimationWaveform 1D-CNN + conformal PIsCaveat: Nominal-90 % split-conformal coverage 26.5 % at SBP ≥ 160 and 68.9 % on MIMIC-III external; DBP not transcribed.13.0MAE (mmHg), lower is better#2 of 2

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 pressure

Signals

PPG

Keywords

cuffless-bpcalibration-freeconformal-predictionprediction-intervaluncertaintyCQRMondriandataset-shiftpatient-levelVitalDBMIMIC-IIILightGBM1D-CNNpre-registered-analysis-plantrends-first

Datasets

How it works

How Method works: Setup → Method → TargetSetup: Signals (PPG); Form factor (clinical_finger). Method: Method (Convolutional network · Gradient-boosted trees). Target: Blood pressure.SetupSignals: PPGSignalsPPGForm factor: clinical_fingerForm factorclinical_fingerMethodMethodConvolutional network ·…TargetBlood pressureBlood pressure

Cite this paper

APA

Where Do Prediction Intervals Fail for Calibration-Free Cuffless Blood Pressure? A Patient-Level Analysis Within and Across Datasets. (2026). Submitted to Physiological Measurement. https://doi.org/10.5281/zenodo.23513090

BibTeX

@article{fukushimaconform2026where,
  title         = {{Where Do Prediction Intervals Fail for Calibration-Free Cuffless Blood Pressure? A Patient-Level Analysis Within and Across Datasets}},
  year          = {2026},
  journal       = {Submitted to Physiological Measurement},
  doi           = {10.5281/zenodo.23513090}
}

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

Record updated

Added to the ChatPPG research hub