Sound signature for identification and quantification of upper airway disease in horses.

The research looks at whether sound signatures can accurately identify and quantify upper airway diseases such as laryngeal hemiplegia and alar fold paralysis in horses, distinguishing them from healthy cases. The authors found that analysing peak sound level and highest frequency of sound intensity improved the specificity and sensitivity of disease identification with an accuracy rate of 92%.

Methods Employed

• The study was performed on six mature horses, with upper airway sounds recorded under three different conditions: normal upper airway function (as a control), after induction of left laryngeal hemiplegia, or after induction of bilateral alar fold paralysis.
• The horses were exercised on a high-speed treadmill until they reached their maximum heart rate (HRmax), under the various induced conditions.
• The sounds were analysed in terms of fundamental frequency using Gabor transform, and intensity characteristics, specifically peak sound level (soundpeak) and highest frequency of at least -25 dB sound intensity (F(25max)).

Key Findings

• Fundamental frequency of the time domain signal did not prove useful in the sound analysis.
• Spectrograms performed better with sensitivity and specificity of 83% and 75% respectively, but they could correctly identify the exact abnormal condition in only 12 out of 18 spectrograms.
• The accuracy improved dramatically when soundpeak and F(25max) were used as the discriminatory variables. This means that by analyzing the peak sound level and the highest frequency of sound intensity, the researchers were able to better differentiate between control and laryngeal hemiplegia conditions (using soundpeak), and between laryngeal hemiplegia and alar fold paralysis conditions (using F(25max)).
• With this method, the specificity of sound analysis increased to 92%, with the same level of sensitivity (83%). It correctly classified the disease state in 92% of the affected horses.

Conclusion

• The research indicates that sound analysis could potentially be a useful tool in diagnosing and evaluating the treatment of horses with upper airway obstructions. However, close attention needs to be paid to the intensity of exercise.
• Given the limitations of any single measure, the researchers suggest that multiple measurements might be required to attain sufficient accuracy for practical clinical use.