Waveform Similarity Analysis: A Simple Template Comparing Approach for Detecting and Quantifying Noisy Evoked...

Area of Science:

• Neuroscience
• Biomedical Engineering
• Electrophysiology

Background:

• Assessing nerve regeneration electrophysiologically is difficult due to complex, dispersed responses and low signal-to-noise ratios.
• Automating the analysis of evoked potentials is crucial for objective and efficient study of neural repair.

Purpose of the Study:

• To develop and validate an automated algorithm for detecting and quantifying compound action potential events in neural regeneration.
• To compare the performance of the automated algorithm against manual analysis by a trained observer.

Main Methods:

• Developed the Waveform Similarity Analysis (WSA) algorithm using a cross-correlation template matching approach.
• Tested WSA on in vivo evoked potentials from regenerating sural and sciatic nerves in animal models.
• Compared WSA results with manual measurements from a trained observer (Trained Eye Analysis) on simulated and real-world noisy signals.

Main Results:

• WSA successfully detected and quantified compound action potentials from intact and regenerating nerves, including complex waveforms.
• WSA demonstrated objectivity and automation, surpassing manual analysis in detecting small, clustered events with poor signal-to-noise ratios.
• While WSA showed consistent latency errors with imperfect templates in simulations, it proved reliable for amplitude prediction and overall quantification.

Conclusions:

• Waveform Similarity Analysis offers a simple, reliable, and objective method for quantifying evoked potentials in neural regeneration.
• The algorithm is a valuable tool for studying neural repair by overcoming limitations of manual electrophysiological assessment.
• WSA enhances the efficiency and accuracy of analyzing complex waveforms in electrophysiological studies of nerve regeneration.