Machine Learning Using a Single-Lead ECG to Identify Patients With Atrial Fibrillation-Induced Heart Failure

Giorgio Luongo,Louisa Mayer,Franz-Josef Neumann,Deborah Nairn,Björn Müller-Edenborn,Massimo W Rivolta,Axel Loewe,Felix Rees,Amir Jadidi,Olaf Dössel,Roberto Sassi,Christoph Ahlgrim,Thomas Arentz

doi:10.3389/fcvm.2022.812719

Abstract

AimsAtrial fibrillation (AF) and heart failure often co-exist. Early identification of AF patients at risk for AF-induced heart failure (AF-HF) is desirable to reduce both morbidity and mortality as well as health care costs. We aimed to leverage the characteristics of beat-to-beat-patterns in AF to prospectively discriminate AF patients with and without AF-HF.MethodsA dataset of 10,234 5-min length RR-interval time series derived from 26 AF-HF patients and 26 control patients was extracted from single-lead Holter-ECGs. A total of 14 features were extracted, and the most informative features were selected. Then, a decision tree classifier with 5-fold cross-validation was trained, validated, and tested on the dataset randomly split. The derived algorithm was then tested on 2,261 5-min segments from six AF-HF and six control patients and validated for various time segments.ResultsThe algorithm based on the spectral entropy of the RR-intervals, the mean value of the relative RR-interval, and the root mean square of successive differences of the relative RR-interval yielded an accuracy of 73.5%, specificity of 91.4%, sensitivity of 64.7%, and PPV of 87.0% to correctly stratify segments to AF-HF. Considering the majority vote of the segments of each patient, 10/12 patients (83.33%) were correctly classified.ConclusionBeat-to-beat-analysis using a machine learning classifier identifies patients with AF-induced heart failure with clinically relevant diagnostic properties. Application of this algorithm in routine care may improve early identification of patients at risk for AF-induced cardiomyopathy and improve the yield of targeted clinical follow-up.

Highlights

Atrial fibrillation (AF) is the most common arrhythmia in human kind, affecting approximately eight million patients in the European Union [1]
We show the methods and procedures used to implement a classification between patients with AF-HF and control group patients using 5-min RR signals acquired during daylight hours
As the current study focuses on AF-induced heart failure, only patients who experienced an absolute improvement of left ventricular ejection fraction (LVEF) of 15% or more within 40 days in sinus rhythm remained in the study for further analysis [6]

Summary

Introduction

Atrial fibrillation (AF) is the most common arrhythmia in human kind, affecting approximately eight million patients in the European Union [1]. AF can occur concomitantly with heart failure without causative relation, and restoration of sinus rhythm in these patients results in only modest improvements of left ventricular systolic dysfunction (LVSD). In a potentially large subset of patients with AF and heart failure sinus rhythm restoration leads to drastic improvements or normalization of LVSD [3–6] within days to weeks. It is currently not fully understood why certain patients develop severe heart failure symptoms and LVSD during AF (AF-induced heart failure; AF-HF). Given the ever-increasing prevalence of AF in the European population, applicable screening tools to identify patients at risk are desirable to tailor patient care and reduce costs for health care systems

Methods

Results

Discussion

Conclusion