The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Dans de nombreuses situations, des sons anormaux, appelés sons fortuits, sont inclus avec les bruits pulmonaires d'un sujet souffrant de maladies pulmonaires. Ainsi, une méthode pour détecter automatiquement les sons anormaux lors de l’auscultation a été proposée. Les caractéristiques acoustiques des bruits pulmonaires normaux pour les sujets témoins et des bruits pulmonaires anormaux pour les patients sont exprimées à l'aide de modèles de Markov cachés (HMM) pour distinguer les bruits pulmonaires normaux et anormaux. De plus, des sons anormaux ont été détectés dans un environnement bruyant, notamment des bruits cardiaques, à l'aide d'un modèle de bruits cardiaques. Cependant, le score F1 obtenu pour détecter une respiration anormale était faible (0.8493). De plus, la durée et les propriétés acoustiques des segments de bruits respiratoires, cardiaques et fortuits variaient. Dans notre méthode précédente, les HMM appropriés pour les segments cardiaques et sonores adventifs ont été construits. Bien que les propriétés des types de sons fortuits varient, une topologie appropriée pour chaque type n’a pas été prise en compte. Dans cette étude, des HMM appropriés pour les segments de chaque type de son fortuit et d'autres segments ont été construits. Le score F1 a été augmenté (0.8726) en sélectionnant une topologie appropriée pour chaque segment. Les résultats démontrent l'efficacité de la méthode proposée.
Masaru YAMASHITA
Nagasaki University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Masaru YAMASHITA, "Acoustic HMMs to Detect Abnormal Respiration with Limited Training Data" in IEICE TRANSACTIONS on Information,
vol. E106-D, no. 3, pp. 374-380, March 2023, doi: 10.1587/transinf.2022EDP7068.
Abstract: In many situations, abnormal sounds, called adventitious sounds, are included with the lung sounds of a subject suffering from pulmonary diseases. Thus, a method to automatically detect abnormal sounds in auscultation was proposed. The acoustic features of normal lung sounds for control subjects and abnormal lung sounds for patients are expressed using hidden markov models (HMMs) to distinguish between normal and abnormal lung sounds. Furthermore, abnormal sounds were detected in a noisy environment, including heart sounds, using a heart-sound model. However, the F1-score obtained in detecting abnormal respiration was low (0.8493). Moreover, the duration and acoustic properties of segments of respiratory, heart, and adventitious sounds varied. In our previous method, the appropriate HMMs for the heart and adventitious sound segments were constructed. Although the properties of the types of adventitious sounds varied, an appropriate topology for each type was not considered. In this study, appropriate HMMs for the segments of each type of adventitious sound and other segments were constructed. The F1-score was increased (0.8726) by selecting a suitable topology for each segment. The results demonstrate the effectiveness of the proposed method.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2022EDP7068/_p
Copier
@ARTICLE{e106-d_3_374,
author={Masaru YAMASHITA, },
journal={IEICE TRANSACTIONS on Information},
title={Acoustic HMMs to Detect Abnormal Respiration with Limited Training Data},
year={2023},
volume={E106-D},
number={3},
pages={374-380},
abstract={In many situations, abnormal sounds, called adventitious sounds, are included with the lung sounds of a subject suffering from pulmonary diseases. Thus, a method to automatically detect abnormal sounds in auscultation was proposed. The acoustic features of normal lung sounds for control subjects and abnormal lung sounds for patients are expressed using hidden markov models (HMMs) to distinguish between normal and abnormal lung sounds. Furthermore, abnormal sounds were detected in a noisy environment, including heart sounds, using a heart-sound model. However, the F1-score obtained in detecting abnormal respiration was low (0.8493). Moreover, the duration and acoustic properties of segments of respiratory, heart, and adventitious sounds varied. In our previous method, the appropriate HMMs for the heart and adventitious sound segments were constructed. Although the properties of the types of adventitious sounds varied, an appropriate topology for each type was not considered. In this study, appropriate HMMs for the segments of each type of adventitious sound and other segments were constructed. The F1-score was increased (0.8726) by selecting a suitable topology for each segment. The results demonstrate the effectiveness of the proposed method.},
keywords={},
doi={10.1587/transinf.2022EDP7068},
ISSN={1745-1361},
month={March},}
Copier
TY - JOUR
TI - Acoustic HMMs to Detect Abnormal Respiration with Limited Training Data
T2 - IEICE TRANSACTIONS on Information
SP - 374
EP - 380
AU - Masaru YAMASHITA
PY - 2023
DO - 10.1587/transinf.2022EDP7068
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E106-D
IS - 3
JA - IEICE TRANSACTIONS on Information
Y1 - March 2023
AB - In many situations, abnormal sounds, called adventitious sounds, are included with the lung sounds of a subject suffering from pulmonary diseases. Thus, a method to automatically detect abnormal sounds in auscultation was proposed. The acoustic features of normal lung sounds for control subjects and abnormal lung sounds for patients are expressed using hidden markov models (HMMs) to distinguish between normal and abnormal lung sounds. Furthermore, abnormal sounds were detected in a noisy environment, including heart sounds, using a heart-sound model. However, the F1-score obtained in detecting abnormal respiration was low (0.8493). Moreover, the duration and acoustic properties of segments of respiratory, heart, and adventitious sounds varied. In our previous method, the appropriate HMMs for the heart and adventitious sound segments were constructed. Although the properties of the types of adventitious sounds varied, an appropriate topology for each type was not considered. In this study, appropriate HMMs for the segments of each type of adventitious sound and other segments were constructed. The F1-score was increased (0.8726) by selecting a suitable topology for each segment. The results demonstrate the effectiveness of the proposed method.
ER -