PURPOSE: The broad, long term objective of this research is to create an algorithm and then a device to detect pneumothorax (PTX) quickly and under a variety of environmental conditions. The goal in this particular study was to investigate differences in the acoustic patterns of patients with pneumothorax as compared to sound patterns in healthy controls.
METHODS: Using a Multichannel Lung Sound Analyzer (Stethographics STG16) we studied 2 patients with pneumothoraces and 16 patients who had undergone pneumonectomy. We considered pneumonectomy patients to have findings acoustically equivalent to patients with pneumothoraces. Based on observations in these patients we developed an algorithm for the detection of PTX and tested it in 125 healthy subjects and on 50 recordings on pneumonectomy patients.
RESULTS: The algorithm had a sensitivity of 93% and a specificity of 100%.
CONCLUSION: These results demonstrate the possibility of developing an automated PTX detector based on acoustic differences between collapsed and normal lung.
CLINICAL IMPLICATIONS: Automated acoustic analysis has a potential value in diagnosis of PTX.
DISCLOSURE: Raymond Murphy, No Product/Research Disclosure Information; Shareholder R. Murphy is shareholder of Stethographics, Inc.; Employee R. Murphy is an employee of Stethographics, Inc.