Dataset Information

IRIDIA-AF, a large paroxysmal atrial fibrillation long-term electrocardiogram monitoring database.

ABSTRACT: Atrial fibrillation (AF) is the most common sustained heart arrhythmia in adults. Holter monitoring, a long-term 2-lead electrocardiogram (ECG), is a key tool available to cardiologists for AF diagnosis. Machine learning (ML) and deep learning (DL) models have shown great capacity to automatically detect AF in ECG and their use as medical decision support tool is growing. Training these models rely on a few open and annotated databases. We present a new Holter monitoring database from patients with paroxysmal AF with 167 records from 152 patients, acquired from an outpatient cardiology clinic from 2006 to 2017 in Belgium. AF episodes were manually annotated and reviewed by an expert cardiologist and a specialist cardiac nurse. Records last from 19 hours up to 95 hours, divided into 24-hour files. In total, it represents 24 million seconds of annotated Holter monitoring, sampled at 200 Hz. This dataset aims at expanding the available options for researchers and offers a valuable resource for advancing ML and DL use in the field of cardiac arrhythmia diagnosis.

SUBMITTER: Gilon C

PROVIDER: S-EPMC10584890 | biostudies-literature | 2023 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

IRIDIA-AF, a large paroxysmal atrial fibrillation long-term electrocardiogram monitoring database.

Gilon Cédric C Grégoire Jean-Marie JM Mathieu Marianne M Carlier Stéphane S Bersini Hugues H

Scientific data 20231018 1

Atrial fibrillation (AF) is the most common sustained heart arrhythmia in adults. Holter monitoring, a long-term 2-lead electrocardiogram (ECG), is a key tool available to cardiologists for AF diagnosis. Machine learning (ML) and deep learning (DL) models have shown great capacity to automatically detect AF in ECG and their use as medical decision support tool is growing. Training these models rely on a few open and annotated databases. We present a new Holter monitoring database from patients w ...[more]

PMID: 37853076

Similar Datasets

Project description:BackgroundAtrial fibrillation (AFib) is the most common cardiac arrhythmia associated with stroke, blood clots, heart failure, coronary artery disease, and/or death. Multiple methods have been proposed for AFib detection, with varying performances, but no single approach appears to be optimal. We hypothesized that each state-of-the-art algorithm is appropriate for different subsets of patients and provides some independent information. Therefore, a set of suitably chosen algorithms, combined in a weighted voting framework, will provide a superior performance to any single algorithm.MethodsWe investigate and modify 38 state-of-the-art AFib classification algorithms for a single-lead ambulatory electrocardiogram (ECG) monitoring device. All algorithms are ranked using a random forest classifier and an expert-labeled training dataset of 2,532 recordings. The seven top-ranked algorithms are combined by using an optimized weighting approach.ResultsThe proposed fusion algorithm, when validated on a separate test dataset consisting of 4,644 recordings, resulted in an area under the receiver operating characteristic (ROC) curve of 0.99. The sensitivity, specificity, positive-predictive-value (PPV), negative-predictive-value (NPV), and F1-score of the proposed algorithm were 0.93, 0.97, 0.87, 0.99, and 0.90, respectively, which were all superior to any single algorithm or any previously published.ConclusionThis study demonstrates how a set of well-chosen independent algorithms and a voting mechanism to fuse the outputs of the algorithms, outperforms any single state-of-the-art algorithm for AFib detection. The proposed framework is a case study for the general notion of crowdsourcing between open-source algorithms in healthcare applications. The extension of this framework to similar applications may significantly save time, effort, and resources, by combining readily existing algorithms. It is also a step toward the democratization of artificial intelligence and its application in healthcare.

Project description:BackgroundFew data on the thromboembolic (TE) risk of paroxysmal and persistent atrial fibrillation (AF) are available. This study aimed to assess the incidence of TE events in paroxysmal and persistent AF.MethodsWe performed a subset post hoc analysis of 771 patients with paroxysmal and 463 with persistent AF enrolled in the multicenter, prospective, randomized, double-blind, placebo-controlled GISSI-AF trial - comparing the efficacy of valsartan versus placebo in preventing AF recurrences - where the choice of antithrombotic treatment was left to the judgment of the referring physician. TE and major outcome events were centrally validated. AF recurrences were detected by frequent clinic visits and a transtelephonic monitoring device with weekly and symptomatic transmissions.ResultsEighty-five percent of patients had a history of hypertension, and the 7.7% had heart failure, left ventricular dysfunction, or both. The mean CHADS2 score was 1.41±0.84. TE and major bleeding events were observed at a low incidence among the overall population at 1-year follow-up (0.97% and 0.81%, respectively). The univariate and multivariable analyses revealed no statistically significant differences in the incidence of TE, major bleeding events or mortality in paroxysmal and persistent AF patients. TE events were more common among women than men (p=0.02). The follow-up examination showed under- or overtreatment with warfarin in many patients, according to guideline suggestions. Warfarin was more frequently prescribed to patients with persistent AF (p<0.0001) and patients with AF recurrences (p<0.0001). AF recurrences were noninvasively detected in 632 (51.2%) patients. In patients without AF recurrences, the TE event rate was 0.5% versus 1.74%, 1.28%, and 1.18% for those with only symptomatic, only asymptomatic or both symptomatic and asymptomatic AF recurrences, respectively, but the difference was not statistically significant, even after adjusting for warfarin treatment and the CHADS2 score (HR 2.93; CI 95%; 0.8-10.9; p=0.11).ConclusionsTE and major bleeding events showed a very low incidence in the GISSI-AF trial population, despite under- or overtreatment with warfarin in many patients. TE events had a similar rate in paroxysmal and persistent AF.Trial registrationTrial registration numberNCT00376272.

Project description:ImportancePatients with ischemic stroke attributed to large- or small-vessel disease are not considered at high risk for atrial fibrillation (AF), and the AF incidence rate in this population is unknown.ObjectivesTo determine whether long-term cardiac monitoring is more effective than usual care for AF detection in patients with stroke attributed to large- or small-vessel disease through 12 months of follow-up.Design, setting, and participantsThe STROKE-AF trial was a randomized (1:1), multicenter (33 sites in the US) clinical trial that enrolled 496 patients between April 2016 and July 2019, with primary end point follow-up through August 2020. Eligible patients were aged 60 years or older or aged 50 to 59 years with at least 1 additional stroke risk factor and had an index stroke attributed to large- or small-vessel disease within 10 days prior to insertable cardiac monitor (ICM) insertion.InterventionsPatients randomized to the intervention group (n = 242) received ICM insertion within 10 days of the index stroke; patients in the control group (n = 250) received site-specific usual care consisting of external cardiac monitoring, such as 12-lead electrocardiograms, Holter monitoring, telemetry, or event recorders.Main outcomes and measuresIncident AF lasting more than 30 seconds through 12 months.ResultsAmong 492 patients who were randomized (mean [SD] age, 67.1 [9.4] years; 185 [37.6%] women), 417 (84.8%) completed 12 months of follow-up. The median (interquartile range) CHA2DS2-VASc (congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, stroke or transient ischemic attack, vascular disease, age 65 to 74 years, sex category) score was 5 (4-6). AF detection at 12 months was significantly higher in the ICM group vs the control group (27 patients [12.1%] vs 4 patients [1.8%]; hazard ratio, 7.4 [95% CI, 2.6-21.3]; P < .001). Among the 221 patients in the ICM group who received an ICM, 4 (1.8%) had ICM procedure-related adverse events (1 site infection, 2 incision site hemorrhages, and 1 implant site pain).Conclusions and relevanceAmong patients with stroke attributed to large- or small-vessel disease, monitoring with an ICM compared with usual care detected significantly more AF over 12 months. However, further research is needed to understand whether identifying AF in these patients is of clinical importance.Trial registrationClinicalTrials.gov Identifier: NCT02700945.

Dataset Information

IRIDIA-AF, a large paroxysmal atrial fibrillation long-term electrocardiogram monitoring database.

Publications

IRIDIA-AF, a large paroxysmal atrial fibrillation long-term electrocardiogram monitoring database.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets