CT image segmentation for inflamed and fibrotic lungs using a multi-resolution convolutional neural network
Abstract The purpose of this study was to develop a fully-automated segmentation algorithm, robust to various density enhancing lung abnormalities, to facilitate rapid quantitative analysis of computed tomography images. A polymorphic training approach is proposed, in which both specifically labeled...
Guardado en:
| Autores principales: | , , , , , , , , , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
Nature Portfolio
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/67123f2fad0c4e4f98f42b8ab456559b |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| id |
oai:doaj.org-article:67123f2fad0c4e4f98f42b8ab456559b |
|---|---|
| record_format |
dspace |
| spelling |
oai:doaj.org-article:67123f2fad0c4e4f98f42b8ab456559b2021-12-02T14:12:40ZCT image segmentation for inflamed and fibrotic lungs using a multi-resolution convolutional neural network10.1038/s41598-020-80936-42045-2322https://doaj.org/article/67123f2fad0c4e4f98f42b8ab456559b2021-01-01T00:00:00Zhttps://doi.org/10.1038/s41598-020-80936-4https://doaj.org/toc/2045-2322Abstract The purpose of this study was to develop a fully-automated segmentation algorithm, robust to various density enhancing lung abnormalities, to facilitate rapid quantitative analysis of computed tomography images. A polymorphic training approach is proposed, in which both specifically labeled left and right lungs of humans with COPD, and nonspecifically labeled lungs of animals with acute lung injury, were incorporated into training a single neural network. The resulting network is intended for predicting left and right lung regions in humans with or without diffuse opacification and consolidation. Performance of the proposed lung segmentation algorithm was extensively evaluated on CT scans of subjects with COPD, confirmed COVID-19, lung cancer, and IPF, despite no labeled training data of the latter three diseases. Lobar segmentations were obtained using the left and right lung segmentation as input to the LobeNet algorithm. Regional lobar analysis was performed using hierarchical clustering to identify radiographic subtypes of COVID-19. The proposed lung segmentation algorithm was quantitatively evaluated using semi-automated and manually-corrected segmentations in 87 COVID-19 CT images, achieving an average symmetric surface distance of $$0.495\pm 0.309$$ 0.495 ± 0.309 mm and Dice coefficient of $$0.985\pm 0.011$$ 0.985 ± 0.011 . Hierarchical clustering identified four radiographical phenotypes of COVID-19 based on lobar fractions of consolidated and poorly aerated tissue. Lower left and lower right lobes were consistently more afflicted with poor aeration and consolidation. However, the most severe cases demonstrated involvement of all lobes. The polymorphic training approach was able to accurately segment COVID-19 cases with diffuse consolidation without requiring COVID-19 cases for training.Sarah E. GerardJacob HerrmannYi XinKevin T. MartinEmanuele RezoagliDavide IppolitoGiacomo BellaniMaurizio CeredaJunfeng GuoEric A. HoffmanDavid W. KaczkaJoseph M. ReinhardtNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-12 (2021) |
| institution |
DOAJ |
| collection |
DOAJ |
| language |
EN |
| topic |
Medicine R Science Q |
| spellingShingle |
Medicine R Science Q Sarah E. Gerard Jacob Herrmann Yi Xin Kevin T. Martin Emanuele Rezoagli Davide Ippolito Giacomo Bellani Maurizio Cereda Junfeng Guo Eric A. Hoffman David W. Kaczka Joseph M. Reinhardt CT image segmentation for inflamed and fibrotic lungs using a multi-resolution convolutional neural network |
| description |
Abstract The purpose of this study was to develop a fully-automated segmentation algorithm, robust to various density enhancing lung abnormalities, to facilitate rapid quantitative analysis of computed tomography images. A polymorphic training approach is proposed, in which both specifically labeled left and right lungs of humans with COPD, and nonspecifically labeled lungs of animals with acute lung injury, were incorporated into training a single neural network. The resulting network is intended for predicting left and right lung regions in humans with or without diffuse opacification and consolidation. Performance of the proposed lung segmentation algorithm was extensively evaluated on CT scans of subjects with COPD, confirmed COVID-19, lung cancer, and IPF, despite no labeled training data of the latter three diseases. Lobar segmentations were obtained using the left and right lung segmentation as input to the LobeNet algorithm. Regional lobar analysis was performed using hierarchical clustering to identify radiographic subtypes of COVID-19. The proposed lung segmentation algorithm was quantitatively evaluated using semi-automated and manually-corrected segmentations in 87 COVID-19 CT images, achieving an average symmetric surface distance of $$0.495\pm 0.309$$ 0.495 ± 0.309 mm and Dice coefficient of $$0.985\pm 0.011$$ 0.985 ± 0.011 . Hierarchical clustering identified four radiographical phenotypes of COVID-19 based on lobar fractions of consolidated and poorly aerated tissue. Lower left and lower right lobes were consistently more afflicted with poor aeration and consolidation. However, the most severe cases demonstrated involvement of all lobes. The polymorphic training approach was able to accurately segment COVID-19 cases with diffuse consolidation without requiring COVID-19 cases for training. |
| format |
article |
| author |
Sarah E. Gerard Jacob Herrmann Yi Xin Kevin T. Martin Emanuele Rezoagli Davide Ippolito Giacomo Bellani Maurizio Cereda Junfeng Guo Eric A. Hoffman David W. Kaczka Joseph M. Reinhardt |
| author_facet |
Sarah E. Gerard Jacob Herrmann Yi Xin Kevin T. Martin Emanuele Rezoagli Davide Ippolito Giacomo Bellani Maurizio Cereda Junfeng Guo Eric A. Hoffman David W. Kaczka Joseph M. Reinhardt |
| author_sort |
Sarah E. Gerard |
| title |
CT image segmentation for inflamed and fibrotic lungs using a multi-resolution convolutional neural network |
| title_short |
CT image segmentation for inflamed and fibrotic lungs using a multi-resolution convolutional neural network |
| title_full |
CT image segmentation for inflamed and fibrotic lungs using a multi-resolution convolutional neural network |
| title_fullStr |
CT image segmentation for inflamed and fibrotic lungs using a multi-resolution convolutional neural network |
| title_full_unstemmed |
CT image segmentation for inflamed and fibrotic lungs using a multi-resolution convolutional neural network |
| title_sort |
ct image segmentation for inflamed and fibrotic lungs using a multi-resolution convolutional neural network |
| publisher |
Nature Portfolio |
| publishDate |
2021 |
| url |
https://doaj.org/article/67123f2fad0c4e4f98f42b8ab456559b |
| work_keys_str_mv |
AT sarahegerard ctimagesegmentationforinflamedandfibroticlungsusingamultiresolutionconvolutionalneuralnetwork AT jacobherrmann ctimagesegmentationforinflamedandfibroticlungsusingamultiresolutionconvolutionalneuralnetwork AT yixin ctimagesegmentationforinflamedandfibroticlungsusingamultiresolutionconvolutionalneuralnetwork AT kevintmartin ctimagesegmentationforinflamedandfibroticlungsusingamultiresolutionconvolutionalneuralnetwork AT emanuelerezoagli ctimagesegmentationforinflamedandfibroticlungsusingamultiresolutionconvolutionalneuralnetwork AT davideippolito ctimagesegmentationforinflamedandfibroticlungsusingamultiresolutionconvolutionalneuralnetwork AT giacomobellani ctimagesegmentationforinflamedandfibroticlungsusingamultiresolutionconvolutionalneuralnetwork AT mauriziocereda ctimagesegmentationforinflamedandfibroticlungsusingamultiresolutionconvolutionalneuralnetwork AT junfengguo ctimagesegmentationforinflamedandfibroticlungsusingamultiresolutionconvolutionalneuralnetwork AT ericahoffman ctimagesegmentationforinflamedandfibroticlungsusingamultiresolutionconvolutionalneuralnetwork AT davidwkaczka ctimagesegmentationforinflamedandfibroticlungsusingamultiresolutionconvolutionalneuralnetwork AT josephmreinhardt ctimagesegmentationforinflamedandfibroticlungsusingamultiresolutionconvolutionalneuralnetwork |
| _version_ |
1718391787516592128 |