Dataset

Noise Holter Database

258 two-lead clinical Holter ECGs, 6,255 segments with expert 3-level signal-quality labels (Clean / N1 QRS visible / N2 QRS lost) plus arrhythmia labels, 4 Holter devices (PhysioNet 1.0.0, ODC-By 1.0)

A benchmark for automated signal-quality assessment of ambulatory ECG with naturally occurring noise, rather than the synthetic noise of the MIT-BIH Noise Stress Test Database. Recordings come from routine clinical Holter monitoring, anonymised in the clinic. Beats and arrhythmias were first auto-annotated (SyneScope) and then corrected by a cardiologist; quality was labelled per lead.

At a glance
Type
Dataset
Licence
Not reported
Access
Not reported
Subjects
258
Signals
ECG
PPG sites
None
Setting
In the wild, Clinical
Wrist
No
Raw waveforms
Yes
Updated
Nov 18, 2026

Original description

The Noise Holter Database is a publicly available, multi-arrhythmia Holter ECG dataset comprising 258 two-lead recordings from distinct subjects. The primary objective of making this resource available is to facilitate the development of software for filtering out noisy segments in ambulatory recordings. The dataset includes a wide spectrum of spontaneous cardiac arrhythmias—including atrial fibrillation, ventricular tachycardia, AV block, and supraventricular tachycardia—captured under real-world ambulatory conditions with authentic environmental noise. Encompassing approximately 35 hours of data, the database features 6,255 ECG segments. These segments include expert-assigned multi-label signal quality annotations classified into three functional levels: Clean (43.7%), N1 where QRS is identifiable (36.0%), and N2 where QRS is unidentifiable (20.3%). Inter-rater agreement, assessed on a comprehensive validation subset of 2,166 events, yielded a Cohen's kappa of 0.83, demonstrating the robustness of the labeling process. This dataset effectively addresses the critical absence of publicly available benchmarks with authentic, naturally occurring ambulatory noise annotations.

Verbatim fromPhysioNet· Licence: Open Data Commons Attribution License v1.0

Noise Holter Database: ChatPPG research summary card

Key numbers

258
Subjects
3
Reference labels
ECG, AFIB, SQI
200 Hz
Released sampling rate
ECG 200 Hz (SpiderView, 15-bit) and 256 Hz (SpiderFlash, SpiderSAS 15-bit; DMS myPatch 12-bit); exported WFDB headers state 200 Hz

How it works

How Noise Holter Database works: Devices & body sites → Signals recorded → TasksDevices & body sites: Generic Holter ECG (Chest). PPG sites: None. Signals recorded: Noise Holter Database, 258 subjects · 200 Hz, signals ECG. Tasks: Signal quality; AF / rhythm; ECG rhythm.Devices & body sitesGeneric Holter ECG: ChestGeneric Holter ECGChestPPG sites: NonePPG sitesNoneSignals recordedNoise Holter Database258 subjects · 200 HzECGTasksSignal qualitySignal qualityAF / rhythmAF / rhythmECG rhythmECG rhythm

No model in the hub reports results on this dataset yet.

Sampling
ECG 200 Hz (SpiderView, 15-bit) and 256 Hz (SpiderFlash, SpiderSAS 15-bit; DMS myPatch 12-bit); exported WFDB headers state 200 Hz
Reference labels
ecg, afib, sqi
Device roles
Clinical reference (ECG)

Devices· 1

Full photo credits, sources and licences are on each device page.

Cite this dataset

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

Noise Holter Database [Data set]. (n.d.). physionet. https://doi.org/10.13026/8kac-hn03
@misc{noiseholterdatabnoise,
  title         = {{Noise Holter Database}},
  howpublished  = {physionet},
  doi           = {10.13026/8kac-hn03}
}

Cite the ChatPPG record

Cite the dated snapshot 2026-10 for a fixed, reproducible version. Please also cite the original publications listed above.

APA

ChatPPG. (2026). Noise Holter Database: dataset record (Version 2026-10) [Data set]. ChatPPG Research Hub. https://chatppg.com/research/datasets/noise-holter-database

BibTeX
@misc{chatppg_dataset_noise_holter_database,
  author       = {{ChatPPG}},
  title        = {{Noise Holter Database: dataset record}},
  year         = {2026},
  month        = nov,
  version      = {2026-10},
  howpublished = {\url{https://chatppg.com/research/datasets/noise-holter-database}},
  note         = {ChatPPG Research Hub, Snapshot 2026-10, https://chatppg.com/research/snapshots/2026-10}
}
CITATION.cff
cff-version: 1.2.0
message: "If you use this resource, please cite it as below."
type: dataset
title: "Noise Holter Database: dataset record"
authors:
  - name: "ChatPPG"
url: "https://chatppg.com/research/datasets/noise-holter-database"
repository-artifact: "https://chatppg.com/research/snapshots/2026-10"
version: "2026-10"
date-released: "2026-11-18"

History

  1. Record updated