Paperdoi:10.1109/jsen.2026.3738820

Mechanically Decoupled Organic Photodetector-Integrated Wearable PPG Sensor Toward Ambulatory Blood Pressure Monitoring

Venue listed2026Added Nov 19, 2026

  • DS
  • JH
  • KK
  • SK
  • ᄋ
  • MS

Authors: Dae-Geun Song, Jeong Ha Hwang, Ki-Hoon Kim, Sunyoung Kwon, 이동구, Min-Ho Seo

IEEE Sensors Journal (2026; early access)

TL;DR

A wearable PPG sensor that uses a high-gain organic photodetector (ZnO/P3HT:PCBM, EQE 9532% at −2 V) and moves the electronics off the measurement site so the optical module weighs under 25 g. Monte Carlo optics and vibration analysis tuned the skin coupling. A CatBoost model on 55 features tracked BP during high-intensity indoor cycling and, per the abstract, met the standard's error criteria; the abstract gives no subject count or mmHg errors.

  • Form factor: wearable PPG with a separated lightweight optical module (<25 g); measurement site not stated in the abstract

Abstract

The continuous, noninvasive ambulatory blood pressure monitoring is essential for timely cardiovascular risk assessment, yet the conventional wearable photoplethysmography (PPG) sensors suffer from severe motion artifacts caused by physical instability at the sensor–skin interface. Although hardware miniaturization can suppress these inertial artifacts, reducing sensor dimensions inherently degrades the signal-to-noise ratio (SNR). Herein, we report a motion-robust wearable PPG platform that overcomes this tradeoff by integrating a high-gain organic photodetector (OPD) with a mechanically decoupled architecture. The developed ZnO/P3HT:PCBM bulk-heterojunction OPD exhibits an external quantum efficiency (EQE) of 9532% at −2 V under photoconductive operation, ensuring high-fidelity PPG acquisition even under light and miniaturized hardware conditions. By separating the lightweight optical module (<25 g) from the primary electronic instrumentation, we fundamentally minimized mechanical loading at the measurement site. The comprehensive Monte Carlo optical simulations and kinematic vibration analyses optimized the interface design to secure a stable skin-coupling mechanism under dynamic acceleration. Leveraging an artificial intelligence (AI) framework, a CatBoost-based model trained on 55 multidimensional features successfully tracks continuous blood pressure variations during high-intensity indoor cycling. The model satisfies international medical instrumentation standards for error criteria and achieves top-tier performance grades under the evaluated conditions, demonstrating significant potential for the next-generation personalized digital healthcare platforms.

Verbatim from Semantic Scholar (opens in a new tab) · Abstract via Semantic Scholar; © the authors/publisher

Results on ChatPPG leaderboards

All leaderboards

No results from this paper on ChatPPG leaderboards yet. Browse the leaderboards.

Tasks & signals

Tasks

Blood pressureSignal denoising

Signals

Keywords

Cuffless BPOrganic PhotodetectorOPDMotion ArtifactSensor Skin InterfaceCatBoostExerciseHardware

How it works

How Method works: Setup → Method → TargetSetup: Signals (PPG); Form factor (wearable PPG with a separated lightweight optical module (<25 g); measurement site not stated in the abstract). Method: Method. Target: Blood pressure; Signal denoising.SetupSignals: PPGSignalsPPGForm factor: wearable PPG with a separated lightweight optical module (<25 g); measurement site not stated in the abstractForm factorwearable PPG with a separated…MethodMethodTargetBlood pressureBlood pressureSignal denoisingSignal denoising

Cite this paper

Cite this paper with the fields recorded here. Check it against the publisher before submitting.

Mechanically Decoupled Organic Photodetector-Integrated Wearable PPG Sensor Toward Ambulatory Blood Pressure Monitoring. (2026). IEEE Sensors Journal. https://doi.org/10.1109/jsen.2026.3738820
@article{songseomechanica2026mechanically,
  title         = {{Mechanically Decoupled Organic Photodetector-Integrated Wearable PPG Sensor Toward Ambulatory Blood Pressure Monitoring}},
  year          = {2026},
  journal       = {IEEE Sensors Journal},
  doi           = {10.1109/jsen.2026.3738820}
}

Added to the ChatPPG research hub