Inception-Based Network and Multi-Spectrogram Ensemble Applied To Predict Respiratory Anomalies and Lung Diseases

Lam Pham; Huy Phan; Alexander Schindler; Ross Clarence King; Alfred  Mertins; Ian  McLoughlin

doi:10.1109/EMBC46164.2021.9629857

Inception-Based Network and Multi-Spectrogram Ensemble Applied To Predict Respiratory Anomalies and Lung Diseases

Lam Pham, Huy Phan, Alexander Schindler, Ross Clarence King, Alfred Mertins, Ian McLoughlin

Data Science & Artifical Intelligence

Research output: Chapter in Book or Conference Proceedings › Conference Proceedings with Oral Presentation › peer-review

Abstract

This paper presents an inception-based deep neural network for detecting lung diseases using respiratory sound input. Recordings of respiratory sound collected from patients are first transformed into spectrograms where both spectral and temporal information are well represented, in a process referred to as front-end feature extraction. These spectrograms are then fed into the proposed network, in a process referred to as back-end classification, for detecting whether patients suffer from lung-related diseases. Our experiments, conducted over the ICBHI benchmark metadataset of respiratory sound, achieve competitive ICBHI scores of 0.53/0.45 and 0.87/0.85 regarding respiratory anomaly and disease detection, respectively.

Original language	English
Title of host publication	2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)
Pages	235-256
ISBN (Electronic)	978-1-7281-1179-7
DOIs	https://doi.org/10.1109/EMBC46164.2021.9629857
Publication status	Published - Nov 2021

Research Field

Former Research Field - Data Science

Access to Document

10.1109/EMBC46164.2021.9629857

Cite this

@inproceedings{ab1a8bbe4aaa4246ab3069de0802db59,

title = "Inception-Based Network and Multi-Spectrogram Ensemble Applied To Predict Respiratory Anomalies and Lung Diseases",

abstract = "This paper presents an inception-based deep neural network for detecting lung diseases using respiratory sound input. Recordings of respiratory sound collected from patients are first transformed into spectrograms where both spectral and temporal information are well represented, in a process referred to as front-end feature extraction. These spectrograms are then fed into the proposed network, in a process referred to as back-end classification, for detecting whether patients suffer from lung-related diseases. Our experiments, conducted over the ICBHI benchmark metadataset of respiratory sound, achieve competitive ICBHI scores of 0.53/0.45 and 0.87/0.85 regarding respiratory anomaly and disease detection, respectively.",

author = "Lam Pham and Huy Phan and Alexander Schindler and King, {Ross Clarence} and Alfred Mertins and Ian McLoughlin",

year = "2021",

month = nov,

doi = "10.1109/EMBC46164.2021.9629857",

language = "English",

isbn = "978-1-7281-1180-3",

pages = "235--256",

booktitle = "2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)",

}

Inception-Based Network and Multi-Spectrogram Ensemble Applied To Predict Respiratory Anomalies and Lung Diseases. / Pham, Lam; Phan, Huy; Schindler, Alexander et al.
2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC). 2021. p. 235-256.

Research output: Chapter in Book or Conference Proceedings › Conference Proceedings with Oral Presentation › peer-review

TY - GEN

T1 - Inception-Based Network and Multi-Spectrogram Ensemble Applied To Predict Respiratory Anomalies and Lung Diseases

AU - Pham, Lam

AU - Phan, Huy

AU - Schindler, Alexander

AU - King, Ross Clarence

AU - Mertins, Alfred

AU - McLoughlin, Ian

PY - 2021/11

Y1 - 2021/11

N2 - This paper presents an inception-based deep neural network for detecting lung diseases using respiratory sound input. Recordings of respiratory sound collected from patients are first transformed into spectrograms where both spectral and temporal information are well represented, in a process referred to as front-end feature extraction. These spectrograms are then fed into the proposed network, in a process referred to as back-end classification, for detecting whether patients suffer from lung-related diseases. Our experiments, conducted over the ICBHI benchmark metadataset of respiratory sound, achieve competitive ICBHI scores of 0.53/0.45 and 0.87/0.85 regarding respiratory anomaly and disease detection, respectively.

AB - This paper presents an inception-based deep neural network for detecting lung diseases using respiratory sound input. Recordings of respiratory sound collected from patients are first transformed into spectrograms where both spectral and temporal information are well represented, in a process referred to as front-end feature extraction. These spectrograms are then fed into the proposed network, in a process referred to as back-end classification, for detecting whether patients suffer from lung-related diseases. Our experiments, conducted over the ICBHI benchmark metadataset of respiratory sound, achieve competitive ICBHI scores of 0.53/0.45 and 0.87/0.85 regarding respiratory anomaly and disease detection, respectively.

U2 - 10.1109/EMBC46164.2021.9629857

DO - 10.1109/EMBC46164.2021.9629857

M3 - Conference Proceedings with Oral Presentation

SN - 978-1-7281-1180-3

SP - 235

EP - 256

BT - 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)

ER -

Inception-Based Network and Multi-Spectrogram Ensemble Applied To Predict Respiratory Anomalies and Lung Diseases

Abstract

Research Field

Access to Document

Fingerprint

Cite this