ChatPPG.
Paper

Compact PPG-Based Blood Pressure Estimation Using Hardware-Efficient Knowledge Distillation

Neil Joshua Limbaga, Osama Amin, Basem Shihada, Khaled Nabil Salama (KAUST)
IEEE Sensors Letters2026PPG
Read paper Cite
Published articlePublication date: Sep 2026Added to atlas: Oct 11, 2026Source
Citations0Crossref referencesChecked Oct 16, 2026
GitHub stars—Not available
HF downloads—Not available

Activity counts measure attention, not model quality. Missing counts are unknown. Publication status follows the linked source; a listed venue alone does not confirm peer review.

Students trained on BP labels + teacher predictions + feature-level supervision from PPG-only or hybrid ECG–PPG teachers. PhysioNet PTT and VitalDB. A 1-depth, 16-channel residual U-Net student (20.7k parameters) keeps accuracy with minimal penalty; fits a VCU709 FPGA; nRF5340 MCU deployment is RAM-limited for full-window inference.

  • Venue: IEEE Sensors Letters
  • Signals: PPG
  • Form factor: clinical_finger
At a glance
Venue
IEEE Sensors Letters
Published
2026
DOI
10.1109/LSENS.2026.3720577 (opens in a new tab)
arXiv
Not reported
PMID
Not reported
Licence
Not reported
Signals
PPG
Form factor
Clinical finger
Datasets
Subjects
Not reported
Code
Not reported
Added
Oct 11, 2026

Abstract

Continuous blood pressure (BP) monitoring from photoplethysmography (PPG) is compelling for wearable health applications, yet deep learning models for cuffless BP estimation are rarely evaluated under edge-deployment constraints. We propose a hardware-oriented knowledge distillation framework for compact PPG-only BP estimation, where deployable students are trained using ground-truth BP labels, teacher predictions, and feature-level supervision from either PPG-only or hybrid ECG–PPG teachers. Evaluated on PhysioNet PTT and VitalDB, knowledge distillation improves or maintains the accuracy–efficiency tradeoff of compact students, especially in datasets with minimal data. A compact 1-depth 16-channel Residual UNet student model maintained a deployable size of 20.7 k parameters with minimal penalty on average error. Hardware evaluation shows that the smallest, yet acceptable, student model fits within the VCU709 FPGA budget, while nRF5340 microcontroller unit deployment remains RAM-limited under full-window inference.

Verbatim fromSemantic Scholar· Abstract via Semantic Scholar; © the authors/publisher

Compact PPG-Based Blood Pressure Estimation Using Hardware-Efficient Knowledge Distillation: ChatPPG research summary card

How it works

How Method works: Setup → Method → TargetSetup: Signals (PPG); Form factor (clinical_finger). Method: Method. Target: Blood pressure.SetupSignals: PPGSignalsPPGForm factor: clinical_fingerForm factorclinical_fingerMethodMethodTargetBlood pressureBlood pressure

Datasets used· 1

Dataset
VitalDB
6,388 subjects · physionet
PPGECGRespABP

Cite

@article{compactkdppgbpfpgamculse2026,
  title = {{Compact PPG-Based Blood Pressure Estimation Using Hardware-Efficient Knowledge Distillation}},
  author = {Neil Joshua Limbaga and Osama Amin and Basem Shihada and Khaled Nabil Salama (KAUST)},
  year = {2026},
  journal = {IEEE Sensors Letters},
  doi = {10.1109/LSENS.2026.3720577}
}

History

  1. Added to the ChatPPG research hub

More on ChatPPG