Related Experiment Videos
A new method to detect crackles in respiratory sounds
L Vannuccini1, M Rossi, G Pasquali
1Department of Chemistry, University of Siena, Italy. vannuccini@unisi.it
Summary
This study introduces an automated method for detecting crackles in digitized respiratory sounds. The approach accurately identifies the start of crackles in lung sounds, achieving high sensitivity and specificity.
Area of Science:
- Medical Technology
- Respiratory Medicine
- Signal Processing
Background:
- Crackles are abnormal respiratory sounds indicative of various lung conditions.
- Accurate detection of crackles is crucial for diagnosis and monitoring.
- Existing methods for crackle detection may lack precision or automation.
Purpose of the Study:
- To develop and validate an automatic method for detecting and analyzing crackles in digitized respiratory sounds.
- To accurately identify the starting point of crackles within lung sound signals.
- To improve the performance of automated respiratory sound analysis.
Main Methods:
- A two-step method was developed involving thresholding the first derivative absolute value (FDAV) of lung sounds.
- A derivative-smoothing filter was used to evaluate the first derivative (FD), preserving signal moments.
- Crackle detection in the 'zone of interest' was based on peak analysis of FDAV within a temporal window (TW).
Main Results:
- The proposed method demonstrated high performance as an automatic crackle detector.
- Achieved a sensitivity of 84% and a specificity of 89% in crackle detection.
- The method is specifically designed to accurately pinpoint the starting point of crackles.
Conclusions:
- The developed automatic method provides an effective means for detecting and analyzing crackles in respiratory sounds.
- This technique offers improved accuracy in identifying the onset of crackles.
- The findings support the utility of this method in clinical and research applications for respiratory sound analysis.