Cargando…
Distinguishing nontuberculous mycobacteria from Mycobacterium tuberculosis lung disease from CT images using a deep learning framework
PURPOSE: To develop and evaluate the effectiveness of a deep learning framework (3D-ResNet) based on CT images to distinguish nontuberculous mycobacterium lung disease (NTM-LD) from Mycobacterium tuberculosis lung disease (MTB-LD). METHOD: Chest CT images of 301 with NTM-LD and 804 with MTB-LD confi...
Autores principales: | , , , , , , , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Springer Berlin Heidelberg
2021
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8205608/ https://www.ncbi.nlm.nih.gov/pubmed/34131803 http://dx.doi.org/10.1007/s00259-021-05432-x |
_version_ | 1783708538916306944 |
---|---|
author | Wang, Li Ding, Wenlong Mo, Yan Shi, Dejun Zhang, Shuo Zhong, Lingshan Wang, Kai Wang, Jigang Huang, Chencui Zhang, Shu Ye, Zhaoxiang Shen, Jun Xing, Zhiheng |
author_facet | Wang, Li Ding, Wenlong Mo, Yan Shi, Dejun Zhang, Shuo Zhong, Lingshan Wang, Kai Wang, Jigang Huang, Chencui Zhang, Shu Ye, Zhaoxiang Shen, Jun Xing, Zhiheng |
author_sort | Wang, Li |
collection | PubMed |
description | PURPOSE: To develop and evaluate the effectiveness of a deep learning framework (3D-ResNet) based on CT images to distinguish nontuberculous mycobacterium lung disease (NTM-LD) from Mycobacterium tuberculosis lung disease (MTB-LD). METHOD: Chest CT images of 301 with NTM-LD and 804 with MTB-LD confirmed by pathogenic microbiological examination were retrospectively collected. The differences between the clinical manifestations of the two diseases were analysed. 3D-ResNet was developed to randomly extract data in an 8:1:1 ratio for training, validating, and testing. We also collected external test data (40 with NTM-LD and 40 with MTB-LD) for external validation of the model. The activated region of interest was evaluated using a class activation map. The model was compared with three radiologists in the test set. RESULT: Patients with NTM-LD were older than those with MTB-LD, patients with MTB-LD had more cough, and those with NTM-LD had more dyspnoea, and the results were statistically significant (p < 0.05). The AUCs of our model on training, validating, and testing datasets were 0.90, 0.88, and 0.86, respectively, while the AUC on the external test set was 0.78. Additionally, the performance of the model was higher than that of the radiologist, and without manual labelling, the model automatically identified lung areas with abnormalities on CT > 1000 times more effectively than the radiologists. CONCLUSION: This study shows the efficacy of 3D-ResNet as a rapid auxiliary diagnostic tool for NTB-LD and MTB-LD. Its use can help provide timely and accurate treatment strategies to patients with these diseases. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s00259-021-05432-x. |
format | Online Article Text |
id | pubmed-8205608 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2021 |
publisher | Springer Berlin Heidelberg |
record_format | MEDLINE/PubMed |
spelling | pubmed-82056082021-06-16 Distinguishing nontuberculous mycobacteria from Mycobacterium tuberculosis lung disease from CT images using a deep learning framework Wang, Li Ding, Wenlong Mo, Yan Shi, Dejun Zhang, Shuo Zhong, Lingshan Wang, Kai Wang, Jigang Huang, Chencui Zhang, Shu Ye, Zhaoxiang Shen, Jun Xing, Zhiheng Eur J Nucl Med Mol Imaging Original Article PURPOSE: To develop and evaluate the effectiveness of a deep learning framework (3D-ResNet) based on CT images to distinguish nontuberculous mycobacterium lung disease (NTM-LD) from Mycobacterium tuberculosis lung disease (MTB-LD). METHOD: Chest CT images of 301 with NTM-LD and 804 with MTB-LD confirmed by pathogenic microbiological examination were retrospectively collected. The differences between the clinical manifestations of the two diseases were analysed. 3D-ResNet was developed to randomly extract data in an 8:1:1 ratio for training, validating, and testing. We also collected external test data (40 with NTM-LD and 40 with MTB-LD) for external validation of the model. The activated region of interest was evaluated using a class activation map. The model was compared with three radiologists in the test set. RESULT: Patients with NTM-LD were older than those with MTB-LD, patients with MTB-LD had more cough, and those with NTM-LD had more dyspnoea, and the results were statistically significant (p < 0.05). The AUCs of our model on training, validating, and testing datasets were 0.90, 0.88, and 0.86, respectively, while the AUC on the external test set was 0.78. Additionally, the performance of the model was higher than that of the radiologist, and without manual labelling, the model automatically identified lung areas with abnormalities on CT > 1000 times more effectively than the radiologists. CONCLUSION: This study shows the efficacy of 3D-ResNet as a rapid auxiliary diagnostic tool for NTB-LD and MTB-LD. Its use can help provide timely and accurate treatment strategies to patients with these diseases. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s00259-021-05432-x. Springer Berlin Heidelberg 2021-06-16 2021 /pmc/articles/PMC8205608/ /pubmed/34131803 http://dx.doi.org/10.1007/s00259-021-05432-x Text en © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2021 This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic. |
spellingShingle | Original Article Wang, Li Ding, Wenlong Mo, Yan Shi, Dejun Zhang, Shuo Zhong, Lingshan Wang, Kai Wang, Jigang Huang, Chencui Zhang, Shu Ye, Zhaoxiang Shen, Jun Xing, Zhiheng Distinguishing nontuberculous mycobacteria from Mycobacterium tuberculosis lung disease from CT images using a deep learning framework |
title | Distinguishing nontuberculous mycobacteria from Mycobacterium tuberculosis lung disease from CT images using a deep learning framework |
title_full | Distinguishing nontuberculous mycobacteria from Mycobacterium tuberculosis lung disease from CT images using a deep learning framework |
title_fullStr | Distinguishing nontuberculous mycobacteria from Mycobacterium tuberculosis lung disease from CT images using a deep learning framework |
title_full_unstemmed | Distinguishing nontuberculous mycobacteria from Mycobacterium tuberculosis lung disease from CT images using a deep learning framework |
title_short | Distinguishing nontuberculous mycobacteria from Mycobacterium tuberculosis lung disease from CT images using a deep learning framework |
title_sort | distinguishing nontuberculous mycobacteria from mycobacterium tuberculosis lung disease from ct images using a deep learning framework |
topic | Original Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8205608/ https://www.ncbi.nlm.nih.gov/pubmed/34131803 http://dx.doi.org/10.1007/s00259-021-05432-x |
work_keys_str_mv | AT wangli distinguishingnontuberculousmycobacteriafrommycobacteriumtuberculosislungdiseasefromctimagesusingadeeplearningframework AT dingwenlong distinguishingnontuberculousmycobacteriafrommycobacteriumtuberculosislungdiseasefromctimagesusingadeeplearningframework AT moyan distinguishingnontuberculousmycobacteriafrommycobacteriumtuberculosislungdiseasefromctimagesusingadeeplearningframework AT shidejun distinguishingnontuberculousmycobacteriafrommycobacteriumtuberculosislungdiseasefromctimagesusingadeeplearningframework AT zhangshuo distinguishingnontuberculousmycobacteriafrommycobacteriumtuberculosislungdiseasefromctimagesusingadeeplearningframework AT zhonglingshan distinguishingnontuberculousmycobacteriafrommycobacteriumtuberculosislungdiseasefromctimagesusingadeeplearningframework AT wangkai distinguishingnontuberculousmycobacteriafrommycobacteriumtuberculosislungdiseasefromctimagesusingadeeplearningframework AT wangjigang distinguishingnontuberculousmycobacteriafrommycobacteriumtuberculosislungdiseasefromctimagesusingadeeplearningframework AT huangchencui distinguishingnontuberculousmycobacteriafrommycobacteriumtuberculosislungdiseasefromctimagesusingadeeplearningframework AT zhangshu distinguishingnontuberculousmycobacteriafrommycobacteriumtuberculosislungdiseasefromctimagesusingadeeplearningframework AT yezhaoxiang distinguishingnontuberculousmycobacteriafrommycobacteriumtuberculosislungdiseasefromctimagesusingadeeplearningframework AT shenjun distinguishingnontuberculousmycobacteriafrommycobacteriumtuberculosislungdiseasefromctimagesusingadeeplearningframework AT xingzhiheng distinguishingnontuberculousmycobacteriafrommycobacteriumtuberculosislungdiseasefromctimagesusingadeeplearningframework |