Automated multi-class ECG arrhythmia detection using VMD and multi-task optimization

Q: What is the clinical evidence from this study?

Population: Cardiac arrhythmias. Intervention: SVMD-MTPSO hybrid framework with LightGBM vs. Other machine learning models. Primary outcome: Classification accuracy.

Q: What does this research mean for the field?

A machine-learning framework combining Variational Mode Decomposition and Multi-task Particle Swarm Optimization achieves highly accurate (up to 99.3%) and efficient multi-class detection of cardiac arrhythmias, including atrial fibrillation, ventricular fibrillation, and ventricular tachycardia. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

Y. Murali Krishna; K. Padma Vasavi; M. Krishna Chaitanya; Jammisetty Yedukondalu; N. Udaya Kumar; Beebi Naseeba

doi:10.1038/s41598-026-44103-5

April 1, 2026Scientific ReportsOpen Access

Automated multi-class ECG arrhythmia detection using VMD and multi-task optimization

View Full Paper

Ask AI

Bookmark

Share

Key result

SVMD-MTPSO hybrid framework using LightGBM achieves 99.3% accuracy in detecting multi-class ECG arrhythmias.

Figures

Why the study?

ECG classification is essential for accurately detecting and tracking heart rhythm disorders.

Key Points

To develop an automated multi-class classification framework for detecting cardiac arrhythmias from ECG signals using signal decomposition and multi-task feature optimization.
Decomposed ECG signals into intrinsic mode functions using Variational Mode Decomposition (VMD) with K = 10 modes, a penalty parameter α tuned between 1000–5000, and a convergence tolerance of 10^-7.
Extracted entropy-based metrics (Shannon, approximate, differential, and Rényi entropy) and higher-order statistics (2nd, 3rd, and 4th order cumulants) across decomposed modes.
Applied Multi-task Particle Swarm Optimization (MT-PSO) with binary masking to select optimal feature subsets across machine learning classifiers.
LightGBM achieved the highest overall classification accuracy (ACC 0.993) on the optimized feature set.
HistGradientBoost (ACC 0.991), XGBoost (ACC 0.990), and ExtraTrees (ACC 0.990) demonstrated high detection performance across atrial fibrillation, ventricular fibrillation, ventricular tachycardia, and normal rhythm.
Feature optimization via MT-PSO decreased model execution time and yielded faster convergence compared to standard single-task PSO.

Structured PICO

P

Population

ECG signals from publicly available databases (MIT-BIH Malignant Ventricular Ectopy database, Creighton University Ventricular Tachyarrhythmia database, MIT-BIH arrhythmia database, and PhysioNet Atrial Fibrillation Database) representing Atrial Fibrillation (AF), Ventricular Fibrillation (VF), Normal Rhythm (NR), and Ventricular Tachycardia (VT).

I

Intervention

Automated multi-class ECG classification framework using Variational Mode Decomposition (VMD) for signal decomposition, Multi-task Particle Swarm Optimization (MT-PSO) for feature selection, and machine learning models (LightGBM, HistGradientBoost, XGBoost, ExtraTrees).

C

Comparator

Existing ECG classification frameworks and reported works.

O

Outcome

Classification accuracy (ACC) for identifying cardiac arrhythmias.

An automated multi-class ECG classification framework using VMD and MT-PSO achieved over 99% accuracy in detecting arrhythmias like AF, VF, and VT, demonstrating potential for reliable computer-aided diagnosis.

Implication

May support automated arrhythmia detection in clinical practice; extends existing ML frameworks for ECG classification.

Abstract

Electrocardiogram (ECG) classification is essential for accurately detecting and tracking heart rhythm disorders. This study proposes a multi-class ECG classification framework for identifying cardiac arrhythmias like Atrial Fibrillation (AF), Ventricular Fibrillation (VF), Normal Rhythm (NR), and Ventricular Tachycardia (VT). The ECG signals were decomposed using Variational Mode Decomposition (VMD), and higher-order statistics as well as entropy-based features were extracted from each mode. Multi-task Particle Swarm Optimization (MT-PSO) was employed to reduce redundant features and enhance the discriminative capability of the dataset. Multiple machine-learning models were evaluated, and optimized feature set led to clear performance improvements. The best results were obtained using LightGBM (ACC 0.993), HistGradientBoost (0.991), XGBoost (0.990), and ExtraTrees (0.990). Execution time also decreased for several models after optimization. Confusion-matrix and ROC analyses confirmed reliable detection across all four cardiac classes, and comparison with reported works shows that the proposed framework offers competitive or improved performance for ECG classification.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Krishna et al. (2026) studied Cardiac arrhythmias. SVMD-MTPSO hybrid framework with LightGBM vs. Other machine learning models was evaluated on Classification accuracy. The proposed SVMD-MTPSO hybrid framework using the LightGBM classifier achieved 99.3% accuracy in detecting multi-class ECG arrhythmias.

synapsesocial.com/papers/69cf5f225a333a821460df9f https://doi.org/10.1038/s41598-026-44103-5

Topics

Persistent AF management

Atrial fibrillation

View Full Paper

Ask AI

Bookmark

Share

Automated multi-class ECG arrhythmia detection using VMD and multi-task optimization

Figures

Key Points

Structured PICO

Implication

Abstract

Discussion

Cite This Study

Topics

Automated multi-class ECG arrhythmia detection using VMD and multi-task optimization

Figures

Key Points

Structured PICO

Implication

Abstract

Discussion

Cite This Study

Keep reading

More related papers

More on Persistent AF management

More from Scientific Reports

Topics

Keep reading

More related papers

More on Persistent AF management

More from Scientific Reports