Influence of averaging method on muscle deoxygenation interpretation during repeated-sprint exercise

Area of Science:

• Exercise Physiology
• Biomedical Engineering
• Sports Science

Background:

• Near-infrared spectroscopy (NIRS) is widely used to assess muscle oxygenation during exercise.
• Inconsistent data processing methods for NIRS signals hinder accurate interpretation and inter-study comparisons.
• Deoxyhemoglobin ([HHb]) concentration changes reflect muscle oxygen utilization during physical activity.

Purpose of the Study:

• To investigate the impact of different data smoothing and peak/nadir determination methods on NIRS-derived [HHb] trends during repeated-sprint exercise.
• To compare the efficacy of arithmetic averaging versus digital filtering for NIRS signal processing.
• To identify the optimal analytical approach for accurately quantifying muscle metabolic responses.

Main Methods:

• Nine male participants performed ten 10-second sprints with 30-second recovery intervals.
• Six distinct analysis methods were applied to the NIRS-derived [HHb] signal, including windowed means, moving averages, and Butterworth filtering.
• Peaks and nadirs were identified using predetermined time points and a rolling approach for each method.

Main Results:

• The Butterworth filter effectively reduced signal noise while preserving data integrity (r = 0.9892).
• Arithmetic means, regardless of window length or smoothing, underestimated the magnitude of peak and nadir [HHb] compared to a rolling mean approach.
• Sprint-induced changes in muscle oxygenation (Δ[HHb]) were underestimated when using simple averaging methods.

Conclusions:

• Digital filtering is recommended over arithmetic averaging for smoothing NIRS data to maintain signal accuracy.
• A rolling approach for determining peaks and nadirs provides a more reliable assessment of muscle oxygenation dynamics.
• Implementing these methods enhances the accurate interpretation of muscle oxygenation trends during repeated-sprint exercise.