Cite this dataset
HSBP-DB [Data set]. (n.d.). mendeley.com. https://doi.org/10.17632/fwx556cgjy.1
@misc{hsbpdbhsbpdb,
title = {{HSBP-DB}},
howpublished = {mendeley.com},
doi = {10.17632/fwx556cgjy.1}
}synchronized PCG + cuff BP for multi-class hypertension (n=50, 400 recordings)
HSBP-DB (Bin Safie et al.): 400 raw uncompressed .wav phonocardiography recordings from 50 adults (32M/18F; age 22.70±2.00), each with concurrent oscillometric cuff BP. PCG via 3M Littmann CORE Digital Stethoscope (Model 8480) at Erb's point, 4 kHz. BP via Contec ABPM50 (ESH-IP 2010, ±3 mmHg) on left upper arm.
HSBP-DB is a novel, annotated, and synchronised database of phonocardiography (PCG) and arterial blood pressure (BP) recordings, developed to address the absence of a purpose-built, openly accessible benchmark for multi-class hypertension staging from cardiac acoustic signals.
The dataset comprises 400 raw, uncompressed .wav PCG recordings acquired from 50 adult participants (32 male, 18 female; age 22.70 ± 2.00 years) under two physiological conditions: resting-baseline (5 sessions per participant) and post-exercise (3 sessions per participant). PCG signals were captured using the 3M Littmann CORE Digital Stethoscope (Model 8480) positioned at Erb's point, sampled at 4,000 Hz. Concurrent ground-truth blood pressure was measured using the Contec ABPM50 oscillometric ambulatory monitor (ESH-IP 2010 validated), applying a sequential reference-first acquisition protocol to ensure temporal alignment between the BP label and the corresponding PCG recording.
Each recording is labelled at the recording level according to the 2017 AHA/ACC four-class blood pressure classification schema (Normal, Elevated, Hypertension Stage 1, Hypertension Stage 2). All metadata — participant identifier, session index, physiological state (resting/post-exercise), systolic pressure, diastolic pressure, and mean arterial pressure — are self-contained within a structured, anonymised filename convention, eliminating dependency on external metadata files.
Files are organised into four class-stratified directories (Normal, Elevated, HTN-1, HTN-2) corresponding to the AHA/ACC classification of each recording. HSBP-DB is intended to support reproducible research in cuffless blood pressure estimation, PCG-based haemodynamic biomarker extraction, ordinal/multi-class classification modelling, and multimodal cardiovascular signal processing.
Ethical approval was granted by the Universiti Kuala Lumpur Research Ethics Committee (UREC Ref: UNIKL REC/2025/FRGS/APPV/01). All data were anonymised at the point of acquisition; no personally identifiable information is contained in or recoverable from the deposited files.
Verbatim fromDataCite· Licence: info:eu-repo/semantics/openAccess

No model in the hub reports results on this dataset yet.
HSBP-DB [Data set]. (n.d.). mendeley.com. https://doi.org/10.17632/fwx556cgjy.1
@misc{hsbpdbhsbpdb,
title = {{HSBP-DB}},
howpublished = {mendeley.com},
doi = {10.17632/fwx556cgjy.1}
}Cite the dated snapshot 2026-10 for a fixed, reproducible version. Please also cite the original publications listed above.
ChatPPG. (2026). HSBP-DB: dataset record (Version 2026-10) [Data set]. ChatPPG Research Hub. https://chatppg.com/research/datasets/hsbp-db
@misc{chatppg_dataset_hsbp_db,
author = {{ChatPPG}},
title = {{HSBP-DB: dataset record}},
year = {2026},
month = nov,
version = {2026-10},
howpublished = {\url{https://chatppg.com/research/datasets/hsbp-db}},
note = {ChatPPG Research Hub, Snapshot 2026-10, https://chatppg.com/research/snapshots/2026-10}
}cff-version: 1.2.0 message: "If you use this resource, please cite it as below." type: dataset title: "HSBP-DB: dataset record" authors: - name: "ChatPPG" url: "https://chatppg.com/research/datasets/hsbp-db" repository-artifact: "https://chatppg.com/research/snapshots/2026-10" version: "2026-10" date-released: "2026-11-10"
Record updated