Related Experiment Video
Updated: May 15, 2026
Multi-Modal Home Sleep Monitoring in Older Adults
Published on: January 26, 2019
Usefulness of lung sound data collection using Skeeper SM-300® device: A pilot study
Sue In Choi1, Yujin Jeong2, Won Jai Jung1
1Division of Respiratory, Allergy, and Critical Care Medicine, Korea University College of Medicine, Korea University Medical Center, Seoul, Republic of Korea.
Background:
Digital stethoscope has been introduced to clinical practice and are developing in many aspects. Overwhelming pandemic requires non-contact healthcare support, and recent growth of artificial intelligence necessitates high-quality data collection.
Objectives:
We studied the usefulness of a novel, remotely applicable device Skeeper SM-300®.
Methods:
A single center, prospective, single-arm, investigator-blinded investigation was conducted. Sixty-four participants (normal individuals, stable/exacerbated COPD, stable/exacerbated bronchial asthma, pneumonia, ILD, and pleural effusion) had their lung sounds examined with a conventional stethoscope, while simultaneously recorded using the device, and classified into normal, crackle, wheezing and decreased sound. Three specialists (Raters) listened to the digitally recorded lung sounds and classified them. Non-inferiority of Skeeper SM-300® compared with conventional stethoscopy was examined. Audible acceptability of tidal breathing sounds compared with deep breathing sound, examinees' comfort with the new device, acceptability of digital lung sounds, and diagnostic accuracy after listening to the recorded lung sounds were also evaluated.
Results:
Among four breathing sounds, wheezing had the highest percentage of correct answers. Three Raters' judgment of classifying electronic lung sounds showed consistent agreement with the conventional stethoscopic result. demonstrating statistically significant agreement between the conventional and electronic stethoscopy (к = 0.8337, к = 0.8158, and к = 0.8043, respectively). Audible acceptability of tidal breathing sounds resulted in fair agreement, with a к-value ranging from 0.5 to 0.8. A majority of patients expressed that they were fairly comfortable with being examined with the device. Diagnostic accuracy on listening to "deep breath" showed a nearly perfect agreement.
Conclusion:
We verified that the novel electronic stethoscope showed high credibility, yielded highly accurate results, and exhibited non-inferiority compared with conventional stethoscopy. Our investigation supports the usefulness of this novel digital auscultatory device as a credible substitute for conventional analogue stethoscopy.
