CardioMark: Open-Source MATLAB Toolbox for ECG Annotation in Machine Learning and Research

Authors

Samir Abdel-Rahman ^{a, b, *}
Pavel Antiperovitch ^c
Anthony Tang ^c
Vijay Parsa ^a
James C. Lacefield ^{a, b, d}

Affiliations

• ^a Department of Electrical & Computer Engineering, Western University, London, ON, Canada
• ^b Robarts Research Institute, Western University, London, ON, Canada
• ^c Department of Medicine, Schulich School of Medicine & Dentistry, Western University, London, ON, Canada
• ^d Department of Medical Biophysics, Western University, London, ON, Canada

Corresponding Author

Name: Samir Abdel-Rahman
Phone: 519-661-2111 ext. 84303
Fax: 519-850-2436

Abstract

With the growing use of machine learning, especially deep learning, in ECG analysis, access to well-annotated databases has become increasingly important. However, most public ECG databases lack detailed QRS morphology labels or complete 12-lead ECG recordings. Manual annotation remains essential but is time-consuming, subjective, and often requires multiple reviewers. CardioMark is an open-access MATLAB toolbox that simplifies and improves ECG annotation through an intuitive graphical interface. It supports multi-observer input, session-based work, and easy export for machine learning applications. Its modular design allows customization for various ECG tasks, making it a useful tool for both research and education in clinical cardiology.

Requirements

This software requires only a basic MATLAB . Tested on MATLAB 2024b

Installation

The software can be run by adding its folder to the MATLAB search path and opening “CardioMark.mlapp” or by installing the “CardioMark V1.0.mlappinstall” file; then the software can be found in the APPS menu in MATLAB.

Documentation & Getting Started

To begin using CardioMark, you can access the user manual in one of the following ways:

• Download CardioMark User Manual V1.0.pdf directly from this repository in [Documentation] .
• Retrieve it via Zenodo using the DOI: 10.5281/zenodo.17122572.

Reference

For further information, please refer to the following article:
S. Abdel-Rahman, P. Antiperovitch, A. Tang, M.I. Daoud, V. Parsa, J.C. Lacefield, Faster R-CNN approach for estimating global QRS duration in electrocardiograms with a limited quantity of annotated data, Comput. Biol. Med. 192 (2025) 110200. DOI: 10.1016/j.compbiomed.2025.110200