Fully Portable Wireless Soft Stethoscope and Machine Learning for Continuous Real-Time Auscultation and Automated Disease Detection

Sung Hoon Lee, Yun Soung Kim, Woon Hong Yeo

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Modern computer-aided auscultation using digital stethoscopes has advantages over traditional acoustic auscultation, but current devices are too bulky for continuous monitoring and suffer from motion artifacts. A new study presents a soft, wireless, imperceptible wearable system that can continuously monitor heart and lung sounds, allowing for more accurate diagnosis of various lung diseases. The system uses machine learning to capture relevant physiological sounds and precisely analyze core mechanics, providing better performance than commercial digital stethoscopes. The soft stethoscope system successfully overcomes the limitations of traditional stethoscopes and commercial digital ones, offering skin-friendly, robust adhesion to the body while minimizing motion artifacts. The system can detect high-quality cardiopulmonary sounds even during daily activities and is capable of diagnosing seven different types of lung diseases with 85% accuracy for seven classes and 96.7% accuracy for normal versus abnormal. The soft stethoscope system's form factor, portability, and high-quality sound recording offer potential applications in sleep studies and biometric security systems. The study concludes that the soft device has the potential for more accurate at-home sleep monitoring and lung disease detection, paving the way for advancements in digital and smart healthcare. Future studies will focus on a large-group clinical trial with the soft stethoscope system to automatically diagnose cardiopulmonary diseases while providing continuous, digital, real-time auscultation. Additionally, integrating the system with other sensing modalities would expand its applications, such as personalized physiological signals for next-generation biometric security systems.

Original languageEnglish
Title of host publicationProceedings - IEEE 73rd Electronic Components and Technology Conference, ECTC 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1433-1437
Number of pages5
ISBN (Electronic)9798350334982
DOIs
StatePublished - 2023
Externally publishedYes
Event73rd IEEE Electronic Components and Technology Conference, ECTC 2023 - Orlando, United States
Duration: 30 May 20232 Jun 2023

Publication series

NameProceedings - Electronic Components and Technology Conference
Volume2023-May
ISSN (Print)0569-5503

Conference

Conference73rd IEEE Electronic Components and Technology Conference, ECTC 2023
Country/TerritoryUnited States
CityOrlando
Period30/05/232/06/23

Keywords

  • Packaging substrates
  • Soft and intelligent packaging
  • flexible
  • flexible/stretchable hybrid electronics
  • pop-up/origami
  • stretchable
  • wearable electronics

Fingerprint

Dive into the research topics of 'Fully Portable Wireless Soft Stethoscope and Machine Learning for Continuous Real-Time Auscultation and Automated Disease Detection'. Together they form a unique fingerprint.

Cite this