Respiratory rate estimation from multilead directions, based on ECG delineation

Area of Science:

• Biomedical Engineering
• Cardiology
• Signal Processing

Background:

• Direct measurement of respiratory rate (Rr) in clinical settings can be challenging.
• Electrocardiogram (ECG) signals contain information that can be leveraged for Rr estimation.
• Existing methods for ECG-based Rr estimation have limitations in accuracy and complexity.

Purpose of the Study:

• To develop and validate a novel method for estimating instantaneous respiratory rate (Rr) from ECG signals.
• To compare the performance of the proposed method against established Rr estimation algorithms.
• To investigate the impact of using Final Directions (FD) derived from ECG spatial loops for Rr estimation.

Main Methods:

• Respiratory rate (Rr) estimation using Final Directions (FD) derived from multi-lead ECG.
• Application of power spectral analysis on spatial loops formed by QRS complex and T wave features.
• Utilized the Physionet MGH/MF dataset for validation.
• Principal Component Analysis (PCA) for orthogonalizing ECG leads.

Main Results:

• The proposed FD-based method achieved a minimum mean absolute error (MAE) of 2.82 bpm for Rr estimation.
• Outperformed frequency tracking (min MAE 4.53 bpm) and Fourier-based methods (min MAE 4.94 bpm) using single ECG leads.
• Superior performance compared to a weighted multi-signal oscillator-based algorithm (min MAE 3.04 bpm) for 2-3 leads.
• Orthogonalized leads using PCA further improved Rr estimation accuracy.

Conclusions:

• The novel method utilizing Final Directions (FD) from ECG spatial loops offers a more accurate approach to estimating respiratory rate (Rr).
• This technique provides a promising non-invasive alternative for respiratory monitoring in clinical environments.
• The integration of Principal Component Analysis enhances the robustness and accuracy of ECG-based Rr estimation.