Acoustic estimation of neck fluid volume

Area of Science:

• Biomedical Engineering
• Respiratory Medicine
• Acoustics

Background:

• Fluid accumulation in the neck narrows the upper airway (UA), increasing collapsibility and potentially worsening obstructive sleep apnea (OSA).
• Current methods for measuring neck fluid volume (NFV) are inconvenient and costly.
• Changes in UA acoustics due to fluid accumulation may offer a new measurement avenue.

Purpose of the Study:

• To develop a novel, non-invasive method for estimating NFV using acoustic measurements.
• To assess the accuracy of this acoustic method in healthy subjects.

Main Methods:

• Twenty-eight healthy subjects were monitored while supine.
• Neck fluid volume (NFV) and tracheal sounds were measured simultaneously using bioimpedance and a microphone.
• Tracheal sound features were analyzed in time and frequency domains, reduced using regression and minimum-redundancy-maximum-relevance methods.
• Feature sets were used with multi-linear regression and a mixture-density neural network to estimate NFV.

Main Results:

• A novel acoustic-based method for non-invasive NFV estimation was successfully developed.
• Relative estimation errors were very low: 1.25% for the regression method and 3.23% for the neural network method.
• The developed technique demonstrated high accuracy in estimating NFV.

Conclusions:

• This novel acoustic-based approach provides a practical and accurate method for non-invasively estimating neck fluid volume.
• The technology holds potential for diagnosing fluid accumulation in the neck.
• It may contribute to understanding the role of neck fluid in the pathogenesis of obstructive sleep apnea (OSA).