A Robust Deep Learning Framework for Detecting Bursts in Muscle Sympathetic Nerve Activity.

Area of Science:

• Physiology
• Neuroscience
• Biomedical Engineering

Background:

• Muscle sympathetic nerve activity (MSNA) is vital for understanding the sympathetic nervous system.
• Accurate detection of MSNA bursts is crucial for quantitative analysis but is labor-intensive.
• Current manual burst detection by experts is prone to errors and burnout.

Purpose of the Study:

• To develop and validate a novel machine learning-based method for automated MSNA burst detection.
• To improve the accuracy and efficiency of analyzing sympathetic nerve activity.
• To provide a robust tool for researchers studying autonomic nervous system function.

Main Methods:

• A convolutional neural network (CNN) model was developed.
• The CNN integrated both integrated MSNA and electrocardiography (ECG) signals.
• The model was trained and validated on expert-annotated data from 41 healthy female participants.

Main Results:

• The machine learning method achieved an average F1 score of 0.87±0.03 in detecting expert-annotated bursts.
• The proposed approach demonstrated superior performance compared to alternative methods.
• The method showed robust burst peak identification in resting autonomic nervous system recordings.

Conclusions:

• The novel machine learning approach offers an accurate and efficient solution for MSNA burst detection.
• This automated method can reduce the burden on expert annotators and minimize errors.
• The findings support the use of AI in advancing the study of sympathetic nerve dynamics.