dart_ecg 0.1.1

Pure Dart port of NeuroKit2 ECG processing module. Provides ECG signal cleaning, R-peak detection, delineation, quality assessment, simulation, and full processing pipeline.

dart_ecg

A Pure Dart port of the popular NeuroKit2 ECG processing module. This library provides a robust, zero-dependency (native Dart) toolkit for Electrocardiogram (ECG) signal processing, including cleaning, R-peak detection, delineation, quality assessment, heart rate calculation, and respiratory rate extraction (EDR).

Features

• ECG Simulation: Generate synthetic ECG signals (ecgsyn method) for testing and development.
• Signal Cleaning: High-performance Butterworth filtering, detrending, and powerline noise removal.
• R-Peak Detection: Robust algorithms (e.g., neurokit) for finding R-peaks even in noisy signals.
• Wave Delineation: Delineate P, Q, S, and T waves using DWT (Discrete Wavelet Transform), CWT (Continuous Wavelet Transform), or Peak methods.
• ECG Quality: Estimate signal quality to reject noisy epochs.
• ECG-Derived Respiration (EDR): Extract respiratory rates from ECG using methods like vangent2019, charlton2016, soni2019, and sarkar2015.
• 100% NeuroKit2 Parity: Numerically validated against the Python NeuroKit2 reference.

Usage

A simple usage example:

import 'package:dart_ecg/dart_ecg.dart';

void main() {
  // 1. Simulate an ECG signal
  final ecg = ecgSimulate(duration: 10, samplingRate: 250, heartRate: 70);

  // 2. Process the signal
  final result = ecgProcess(ecg, samplingRate: 250);

  print('Detected R-Peaks: \${result.rpeakIndices.length}');
  print('Average Heart Rate: \${result.ecgRate.nanmean.toStringAsFixed(1)} BPM');
  
  // 3. Get EDR (ECG-Derived Respiration)
  final edr = ecgRsp(result.ecgRate, samplingRate: 250);
}

Algorithms

This package implements the following algorithms directly in Dart:

• Cleaning: Neurokit, Biosppy, Pantompkins1985, Hamilton2002, Elgendi2010, Engzeemod2012.
• Peak Detection: Neurokit, Pantompkins1985.
• Delineation: DWT (Discrete Wavelet Transform), CWT (Continuous Wavelet Transform), Peak method.
• EDR: Vangent2019, Charlton2016, Soni2019, Sarkar2015.

License

This project is licensed under the MIT License - see the LICENSE file for details. This library is a Dart port of the NeuroKit2 library, which is also licensed under the MIT License.