ECGquant 0.1.3

Processing pipeline for ECG records and quantification of basic and advanced electrocardiographic markers, natively compatible with the PTB and PTBXL databases.

ECGquant

A robust Python processing pipeline for Electrocardiogram (ECG) records, specifically tailored to handle and analyse data from the PTB and PTB-XL databases.

Authorship

Hector Martinez-Navarro (hector.martinez-navarrouv.es),

Ignacio Garcia-Fernandez (ignacio.garciauv.es)

COMMLAB, Universitat de Valencia

Overview

ECGquant automates the extraction, processing, and visualization of electrocardiographic features. Built on top of standard scientific libraries, it provides a reliable and clean interface for clinical data analysis, precise wave delineation, and biomarker quantification.

Features

• Extensively validated
• Disease-agnostic
• Compatible with Physionet databases: native support for loading and parsing PTB and PTB-XL database records via wfdb.
• Signal processing: advanced noise filtering and baseline wander removal utilising scipy and numpy.
• Wave delineation: accurate detection and localization of P, Q, R, S, and T wave peaks, onsets, and offsets.
• Clinical markers: automated identification of critical cardiac markers, including the J-point and the ST segment (isoelectric line).
• Advanced pattern recognition: fQRS detection, tombstone patterns, biphasic T waves, T wave inversion...
• Data management: export, manipulate, and analyse structured patient datasets seamlessly with pandas.
• Visualisation: built-in plotting tools via matplotlib to inspect clean signals and verify extracted fiducial points.¨

Currently, the library provides lead-derived markers. Future versions will incorporate new complex clinically-relevant ECG markers comprehending information from multimple leads.

Installation

You can install the package directly from PyPI:

pip install ECGquant

ECGquant.py provides a working pipeline for processing ECG data from PTB or PTB-XL databases. Make sure you update the path of the database downloaded and decompressed in your computer.

Requirements

The library requires Python >= 3.10 and depends on the following core packages:

• numpy
• scipy
• pandas
• matplotlib
• wfdb

License

This project is licensed under the MIT License. See the LICENSE file for details.