Tracheal opening discrimination during intubation using acoustic features and Gaussian mixture model

Wei Hao Chen, Yu Hsien Chiu, Hsien Chang Wang, Yu Wei Hung, Hao Po Su, Kuo Sheng Cheng

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Correct identification of glottic opening is crucial during endotracheal tube intubation for airway management. Direct glottis visualization by the physician is considered best standard practice, but it is dependent on conditions, skill, and experience. This study proposes an improved oxygen insufflation system that applies an acoustic modeling approach to discriminate glottic and non-glottic areas utilizing the acoustic response from a steady directional airflow at the hypopharynx. An electric stethoscope is placed at the suprasternal notch to record sound produced by insufflation during intubation. The Gaussian mixture model with mel-frequency cepstral coefficients (MFCCs) is used to determine differences between glottic and non-glottic areas. A dataset containing 56.567 seconds of non-glottic sound and 46.472 seconds of glottic sound was recorded from 8 anesthetized adults receiving intubation. Short-time analysis and several objective evaluations were performed to investigate system performance. The evaluation results show that the system achieved a high classification accuracy of 93.24%. The proposed approach outperformed a baseline linear discriminant analysis method with various configurations of linear prediction coding coefficients and MFCCs, and shows potential in improving glottic identification during endotracheal tube intubation.

Original languageEnglish
Pages (from-to)605-611
Number of pages7
JournalJournal of Medical and Biological Engineering
Volume34
Issue number6
DOIs
Publication statusPublished - 2014 Jan 1

Fingerprint

Acoustics
Acoustic waves
Discriminant analysis
Identification (control systems)
Visualization
Oxygen

All Science Journal Classification (ASJC) codes

  • Biomedical Engineering

Cite this

@article{7ba7a93fa17b47e6bca7b6513ad03ba9,
title = "Tracheal opening discrimination during intubation using acoustic features and Gaussian mixture model",
abstract = "Correct identification of glottic opening is crucial during endotracheal tube intubation for airway management. Direct glottis visualization by the physician is considered best standard practice, but it is dependent on conditions, skill, and experience. This study proposes an improved oxygen insufflation system that applies an acoustic modeling approach to discriminate glottic and non-glottic areas utilizing the acoustic response from a steady directional airflow at the hypopharynx. An electric stethoscope is placed at the suprasternal notch to record sound produced by insufflation during intubation. The Gaussian mixture model with mel-frequency cepstral coefficients (MFCCs) is used to determine differences between glottic and non-glottic areas. A dataset containing 56.567 seconds of non-glottic sound and 46.472 seconds of glottic sound was recorded from 8 anesthetized adults receiving intubation. Short-time analysis and several objective evaluations were performed to investigate system performance. The evaluation results show that the system achieved a high classification accuracy of 93.24{\%}. The proposed approach outperformed a baseline linear discriminant analysis method with various configurations of linear prediction coding coefficients and MFCCs, and shows potential in improving glottic identification during endotracheal tube intubation.",
author = "Chen, {Wei Hao} and Chiu, {Yu Hsien} and Wang, {Hsien Chang} and Hung, {Yu Wei} and Su, {Hao Po} and Cheng, {Kuo Sheng}",
year = "2014",
month = "1",
day = "1",
doi = "10.5405/jmbe.1714",
language = "English",
volume = "34",
pages = "605--611",
journal = "Journal of Medical and Biological Engineering",
issn = "1609-0985",
publisher = "Biomedical Engineering Society",
number = "6",

}

Tracheal opening discrimination during intubation using acoustic features and Gaussian mixture model. / Chen, Wei Hao; Chiu, Yu Hsien; Wang, Hsien Chang; Hung, Yu Wei; Su, Hao Po; Cheng, Kuo Sheng.

In: Journal of Medical and Biological Engineering, Vol. 34, No. 6, 01.01.2014, p. 605-611.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Tracheal opening discrimination during intubation using acoustic features and Gaussian mixture model

AU - Chen, Wei Hao

AU - Chiu, Yu Hsien

AU - Wang, Hsien Chang

AU - Hung, Yu Wei

AU - Su, Hao Po

AU - Cheng, Kuo Sheng

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Correct identification of glottic opening is crucial during endotracheal tube intubation for airway management. Direct glottis visualization by the physician is considered best standard practice, but it is dependent on conditions, skill, and experience. This study proposes an improved oxygen insufflation system that applies an acoustic modeling approach to discriminate glottic and non-glottic areas utilizing the acoustic response from a steady directional airflow at the hypopharynx. An electric stethoscope is placed at the suprasternal notch to record sound produced by insufflation during intubation. The Gaussian mixture model with mel-frequency cepstral coefficients (MFCCs) is used to determine differences between glottic and non-glottic areas. A dataset containing 56.567 seconds of non-glottic sound and 46.472 seconds of glottic sound was recorded from 8 anesthetized adults receiving intubation. Short-time analysis and several objective evaluations were performed to investigate system performance. The evaluation results show that the system achieved a high classification accuracy of 93.24%. The proposed approach outperformed a baseline linear discriminant analysis method with various configurations of linear prediction coding coefficients and MFCCs, and shows potential in improving glottic identification during endotracheal tube intubation.

AB - Correct identification of glottic opening is crucial during endotracheal tube intubation for airway management. Direct glottis visualization by the physician is considered best standard practice, but it is dependent on conditions, skill, and experience. This study proposes an improved oxygen insufflation system that applies an acoustic modeling approach to discriminate glottic and non-glottic areas utilizing the acoustic response from a steady directional airflow at the hypopharynx. An electric stethoscope is placed at the suprasternal notch to record sound produced by insufflation during intubation. The Gaussian mixture model with mel-frequency cepstral coefficients (MFCCs) is used to determine differences between glottic and non-glottic areas. A dataset containing 56.567 seconds of non-glottic sound and 46.472 seconds of glottic sound was recorded from 8 anesthetized adults receiving intubation. Short-time analysis and several objective evaluations were performed to investigate system performance. The evaluation results show that the system achieved a high classification accuracy of 93.24%. The proposed approach outperformed a baseline linear discriminant analysis method with various configurations of linear prediction coding coefficients and MFCCs, and shows potential in improving glottic identification during endotracheal tube intubation.

UR - http://www.scopus.com/inward/record.url?scp=84920945746&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84920945746&partnerID=8YFLogxK

U2 - 10.5405/jmbe.1714

DO - 10.5405/jmbe.1714

M3 - Article

AN - SCOPUS:84920945746

VL - 34

SP - 605

EP - 611

JO - Journal of Medical and Biological Engineering

JF - Journal of Medical and Biological Engineering

SN - 1609-0985

IS - 6

ER -