Assessment of the atrial fibrillation burden in Holter electrocardiogram recordings using artificial intelligence

Hennings, Elisa; Coslovsky, Michael; Paladini, Rebecca E.; Aeschbacher, Stefanie; Knecht, Sven; Schlageter, Vincent; Krisai, Philipp; Badertscher, Patrick; Sticherling, Christian; Oßwald, Stefan; Kühne, Michael; Zuern, Christine S.

doi:10.1016/j.cvdhj.2023.01.003

Cited by 6 publications

(7 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ribeiro et al (2020) developed an AI approach that analyzed standard 12‐lead ECGs using a pretrained CNN, which attained expert‐level accuracy in diagnosing AF and other ECG abnormalities. Hennings et al (2023) developed a DL method for AF detection from Holter ECG recordings that showed high correlation and agreement with the assessment performed by physicians. Petroni et al (2023) presented MUSE, a novel technique for AF detection relying on principal component analysis of sub‐beat ECG signals from one or more leads, classified using deep learning.…”

Section: Resultsmentioning

confidence: 89%

Artificial intelligence for atrial fibrillation detection, prediction, and treatment: A systematic review of the last decade (2013–2023)

Salvi,

Acharya,

Seoni

et al. 2024

WIREs Data Min & Knowl

View full text Add to dashboard Cite

Atrial fibrillation (AF) affects more than 30 million individuals worldwide, making it the most prevalent cardiac arrhythmia on a global scale. This systematic review summarizes recent advancements in applying artificial intelligence (AI) techniques for AF detection, prediction, and guiding treatment selection and risk stratification. In adherence with the PRISMA guidelines (Preferred Reporting Items for Systematic Reviews and Meta‐Analyses), a total of 171 pertinent studies conducted between 2013 and 2023 were examined. Studies applying machine learning (ML) and deep learning (DL) to electrocardiogram (ECG), photoplethysmography (PPG), wearable data, and other sources were evaluated. For AF detection, most works employed DL (48 studies) and ML (28 studies) on ECG data. DL methods directly analyzed ECG waveforms and outperformed approaches relying on hand‐crafted features. For prediction and risk stratification, 22 studies used ML while 7 leveraged DL on ECG. An emerging trend showed the growing potential of PPG for AF screening. Overall, AI demonstrated promising capabilities across various AF‐related tasks. However, real‐world implementation faces challenges including a lack of interpretability, the need for multimodal data integration, prospective performance validation, and regulatory compliance. Future research directions involve quantifying model uncertainty, enhancing transparency, and conducting population‐based clinical trials to facilitate translation into practice.This article is categorized under: Application Areas > Health Care Application Areas > Science and Technology Technologies > Artificial Intelligence

show abstract

Section: Resultsmentioning

confidence: 89%

Artificial intelligence for atrial fibrillation detection, prediction, and treatment: A systematic review of the last decade (2013–2023)

Salvi,

Acharya,

Seoni

et al. 2024

WIREs Data Min & Knowl

View full text Add to dashboard Cite

show abstract

“…These include support vector machines, K‐nearest neighbors, and principal component analysis, and so on. 3 , 6 , 10 Despite this, CNN has always shown high efficiency in ECG analysis, with advantages including feature extraction, translation invariance, scale and distortion tolerance, and other aspects. Indeed, PVSC/PVC‐related research has been insufficient.…”

Section: Discussionmentioning

confidence: 99%

“…Although valuable information can be captured, this limitation has been obviously. 6,7 ECG provides valuable information about heart rhythm and structure. However, ECG has limitations in detecting arrhythmias that occur infrequently or unpredictably.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The diagnostic efficiency of artificial intelligence based 2 hours Holter monitoring in premature ventricular and supraventricular contractions detection

Huang,

Fan,

Wang

et al. 2024

Clinical Cardiology

View full text Add to dashboard Cite

BackgroundElectrocardiography (ECG) and 24 hours Holter monitoring (24 h‐Holter) provided valuable information for premature ventricular and supraventricular contractions (PVC and PSVC). Currently, artificial intelligence (AI) based 2 hours single‐lead Holter (2 h‐Holter) monitoring may provide an improved strategy for PSVC/PVC diagnosis.HypothesisAI combined with single‐lead Holter monitoring improves PSVC/PVC detection.MethodsIn total, 170 patients were enrolled between August 2022 and 2023. All patients wore both devices simultaneously; then, we compared diagnostic efficiency, including the sensitivity/specificity/positive predictive‐value (PPV) and negative predictive‐value (NPV) in detecting PSVC/PVC by 24 h‐Holter and 2 h‐Holter.ResultsThe PPV and NPV in patients underwent 2 h‐Holter were 76.00%/87.50% and 96.35%/98.55, respectively, and the sensitivity and specificity were 79.17%/91.30%, and 95.65%/97.84% in PSVC/PVC detection compared with 24 h‐Holter. The areas under the ROC curves (AUCs) for PSVC and PVC were 0.885 and 0.741, respectively (p < .0001).ConclusionsThe potential advantages of the 2 h‐Holter were shortened wearing period, improved convenience, and excellent consistency of diagnosis.

show abstract

“… 3 , 4 Automated algorithms using AI have previously been successfully used to analyze photoplethysmography-based recordings as well as Holter ECGs, non–commercially available wearable ECG recordings, or implantable loop recorder ECGs. 6 , 7 , 8 , 9 , 10 However, limited experience exists regarding the application of an AI algorithm for single-lead ECG classification recorded from different smart devices. To overcome this challenge, we hypothesized that a novel, commercially available AI algorithm (PulseAI; PulseAI Ltd, Belfast, United Kingdom) 11 , 12 allows accurate rhythm detection with fewer inconclusive classifications and with similar or better sensitivity and specificity compared to the automated manufacturer-based SL-ECG classification from different smart devices.…”

Section: Introductionmentioning

confidence: 99%

Reducing the burden of inconclusive smart device single-lead ECG tracings via a novel artificial intelligence algorithm

Weidlich,

Mannhart,

Kennedy

et al. 2024

Cardiovascular Digital Health Journal

Self Cite

View full text Add to dashboard Cite

Assessment of the atrial fibrillation burden in Holter electrocardiogram recordings using artificial intelligence

Cited by 6 publications

References 21 publications

Artificial intelligence for atrial fibrillation detection, prediction, and treatment: A systematic review of the last decade (2013–2023)

Artificial intelligence for atrial fibrillation detection, prediction, and treatment: A systematic review of the last decade (2013–2023)

The diagnostic efficiency of artificial intelligence based 2 hours Holter monitoring in premature ventricular and supraventricular contractions detection

Reducing the burden of inconclusive smart device single-lead ECG tracings via a novel artificial intelligence algorithm

Contact Info

Product

Resources

About