Dataset Information

Discriminating electrocardiographic responses to His-bundle pacing using machine learning.

ABSTRACT: Background:His-bundle pacing (HBP) has emerged as an alternative to conventional ventricular pacing because of its ability to deliver physiological ventricular activation. Pacing at the His bundle produces different electrocardiographic (ECG) responses: selective His-bundle pacing (S-HBP), non-selective His bundle pacing (NS-HBP), and myocardium-only capture (MOC). These 3 capture types must be distinguished from each other, which can be challenging and time-consuming even for experts. Objective:The purpose of this study was to use artificial intelligence (AI) in the form of supervised machine learning using a convolutional neural network (CNN) to automate HBP ECG interpretation. Methods:We identified patients who had undergone HBP and extracted raw 12-lead ECG data during S-HBP, NS-HBP, and MOC. A CNN was trained, using 3-fold cross-validation, on 75% of the segmented QRS complexes labeled with their capture type. The remaining 25% was kept aside as a testing dataset. Results:The CNN was trained with 1297 QRS complexes from 59 patients. Cohen kappa for the neural network's performance on the 17-patient testing set was 0.59 (95% confidence interval 0.30 to 0.88; P <.0001), with an overall accuracy of 75%. The CNN's accuracy in the 17-patient testing set was 67% for S-HBP, 71% for NS-HBP, and 84% for MOC. Conclusion:We demonstrated proof of concept that a neural network can be trained to automate discrimination between HBP ECG responses. When a larger dataset is trained to higher accuracy, automated AI ECG analysis could facilitate HBP implantation and follow-up and prevent complications resulting from incorrect HBP ECG analysis.

SUBMITTER: Arnold AD

PROVIDER: S-EPMC7484933 | biostudies-literature | 2020 Jul-Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Discriminating electrocardiographic responses to His-bundle pacing using machine learning.

Arnold Ahran D AD Howard James P JP Gopi Aiswarya A AA Chan Cheng Pou CP Ali Nadine N Keene Daniel D Shun-Shin Matthew J MJ Ahmad Yousif Y Wright Ian J IJ Ng Fu Siong FS Linton Nick W F NWF Kanagaratnam Prapa P Peters Nicholas S NS Rueckert Daniel D Francis Darrel P DP Whinnett Zachary I ZI

Cardiovascular digital health journal 20200701 1

<h4>Background</h4>His-bundle pacing (HBP) has emerged as an alternative to conventional ventricular pacing because of its ability to deliver physiological ventricular activation. Pacing at the His bundle produces different electrocardiographic (ECG) responses: selective His-bundle pacing (S-HBP), non-selective His bundle pacing (NS-HBP), and myocardium-only capture (MOC). These 3 capture types must be distinguished from each other, which can be challenging and time-consuming even for experts.<h ...[more]

PMID: 32954375

Similar Datasets

Project description:BackgroundAlthough right ventricular pacing (RVP) is recommended by most of the guidelines for atrioventricular block, it can cause electrical and mechanical desynchrony, impair left ventricular function, and increase the risk of atrial fibrillation. Recently, the His-Purkinje system pacing, including His bundle pacing (HBP) and left bundle branch pacing (LBBP), has emerged as a physiological pacing modality. However, few studies have compared their efficacy and safety in atrioventricular block (AVB).Methods and resultsThe PubMed, Web of Science, Cochrane Library, and ScienceDirect databases were searched for observational studies and randomized trials of patients with atrioventricular block requiring permanent pacing, from database inception until 10 January 2022. The primary outcomes were complications and heart failure hospitalization. The secondary outcomes included changes in left ventricular ejection fraction (LVEF) and left ventricular end-diastolic diameter (LVEDD), pacing parameters, procedure duration, and success rate. After extracting the data at baseline and the longest follow-up duration available, a pairwise meta-analysis and a Bayesian random-effects network meta-analysis were performed. Odds ratios (ORs) with 95% confidence intervals (CIs) or 95% credible intervals (CrIs) were calculated for dichotomous outcomes, whereas mean differences (MDs) with 95% CIs or 95% CrIs were calculated for continuous outcomes. Seven studies and 1,069 patients were included. Overall, 43.4% underwent LBBP, 33.5% HBP, and 23.1% RVP. Compared with RVP, LBBP and HBP were associated with a shorter paced QRS duration and a more preserved LVEF. HBP significantly increased the pacing threshold and reduced the R-wave amplitude. There was no difference in the risk of complications or the implant success rate. The pacing threshold remained stable during follow-up for the three pacing modalities. The pacing impedance was significantly reduced in HBP, while a numerical but non-significant pacing impedance decrease was observed in both LBBP and RVP. LBBP was associated with an increased R-wave amplitude during follow-up.ConclusionIn this systematic review and network meta-analysis, HBP and LBBP were superior to RVP in paced QRS duration and preservation of LVEF for patients with atrioventricular block. LBBP was associated with a lower pacing threshold and a greater R-wave amplitude than HBP. However, the stability of the pacing output of LBBP may be a concern. Further investigation of the long-term efficacy in left ventricular function and the risk of heart failure hospitalization is needed.Systematic review registration[https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=315046], identifier [CRD42022315046].

Dataset Information

Discriminating electrocardiographic responses to His-bundle pacing using machine learning.

Publications

Discriminating electrocardiographic responses to His-bundle pacing using machine learning.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets