Cargando…
A fast and fully-automated deep-learning approach for accurate hemorrhage segmentation and volume quantification in non-contrast whole-head CT
This project aimed to develop and evaluate a fast and fully-automated deep-learning method applying convolutional neural networks with deep supervision (CNN-DS) for accurate hematoma segmentation and volume quantification in computed tomography (CT) scans. Non-contrast whole-head CT scans of 55 pati...
Autores principales: | , , , , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Nature Publishing Group UK
2020
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7652921/ https://www.ncbi.nlm.nih.gov/pubmed/33168895 http://dx.doi.org/10.1038/s41598-020-76459-7 |
_version_ | 1783607796270366720 |
---|---|
author | Arab, Ali Chinda, Betty Medvedev, George Siu, William Guo, Hui Gu, Tao Moreno, Sylvain Hamarneh, Ghassan Ester, Martin Song, Xiaowei |
author_facet | Arab, Ali Chinda, Betty Medvedev, George Siu, William Guo, Hui Gu, Tao Moreno, Sylvain Hamarneh, Ghassan Ester, Martin Song, Xiaowei |
author_sort | Arab, Ali |
collection | PubMed |
description | This project aimed to develop and evaluate a fast and fully-automated deep-learning method applying convolutional neural networks with deep supervision (CNN-DS) for accurate hematoma segmentation and volume quantification in computed tomography (CT) scans. Non-contrast whole-head CT scans of 55 patients with hemorrhagic stroke were used. Individual scans were standardized to 64 axial slices of 128 × 128 voxels. Each voxel was annotated independently by experienced raters, generating a binary label of hematoma versus normal brain tissue based on majority voting. The dataset was split randomly into training (n = 45) and testing (n = 10) subsets. A CNN-DS model was built applying the training data and examined using the testing data. Performance of the CNN-DS solution was compared with three previously established methods. The CNN-DS achieved a Dice coefficient score of 0.84 ± 0.06 and recall of 0.83 ± 0.07, higher than patch-wise U-Net (< 0.76). CNN-DS average running time of 0.74 ± 0.07 s was faster than PItcHPERFeCT (> 1412 s) and slice-based U-Net (> 12 s). Comparable interrater agreement rates were observed between “method-human” vs. “human–human” (Cohen’s kappa coefficients > 0.82). The fully automated CNN-DS approach demonstrated expert-level accuracy in fast segmentation and quantification of hematoma, substantially improving over previous methods. Further research is warranted to test the CNN-DS solution as a software tool in clinical settings for effective stroke management. |
format | Online Article Text |
id | pubmed-7652921 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2020 |
publisher | Nature Publishing Group UK |
record_format | MEDLINE/PubMed |
spelling | pubmed-76529212020-11-12 A fast and fully-automated deep-learning approach for accurate hemorrhage segmentation and volume quantification in non-contrast whole-head CT Arab, Ali Chinda, Betty Medvedev, George Siu, William Guo, Hui Gu, Tao Moreno, Sylvain Hamarneh, Ghassan Ester, Martin Song, Xiaowei Sci Rep Article This project aimed to develop and evaluate a fast and fully-automated deep-learning method applying convolutional neural networks with deep supervision (CNN-DS) for accurate hematoma segmentation and volume quantification in computed tomography (CT) scans. Non-contrast whole-head CT scans of 55 patients with hemorrhagic stroke were used. Individual scans were standardized to 64 axial slices of 128 × 128 voxels. Each voxel was annotated independently by experienced raters, generating a binary label of hematoma versus normal brain tissue based on majority voting. The dataset was split randomly into training (n = 45) and testing (n = 10) subsets. A CNN-DS model was built applying the training data and examined using the testing data. Performance of the CNN-DS solution was compared with three previously established methods. The CNN-DS achieved a Dice coefficient score of 0.84 ± 0.06 and recall of 0.83 ± 0.07, higher than patch-wise U-Net (< 0.76). CNN-DS average running time of 0.74 ± 0.07 s was faster than PItcHPERFeCT (> 1412 s) and slice-based U-Net (> 12 s). Comparable interrater agreement rates were observed between “method-human” vs. “human–human” (Cohen’s kappa coefficients > 0.82). The fully automated CNN-DS approach demonstrated expert-level accuracy in fast segmentation and quantification of hematoma, substantially improving over previous methods. Further research is warranted to test the CNN-DS solution as a software tool in clinical settings for effective stroke management. Nature Publishing Group UK 2020-11-09 /pmc/articles/PMC7652921/ /pubmed/33168895 http://dx.doi.org/10.1038/s41598-020-76459-7 Text en © The Author(s) 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. |
spellingShingle | Article Arab, Ali Chinda, Betty Medvedev, George Siu, William Guo, Hui Gu, Tao Moreno, Sylvain Hamarneh, Ghassan Ester, Martin Song, Xiaowei A fast and fully-automated deep-learning approach for accurate hemorrhage segmentation and volume quantification in non-contrast whole-head CT |
title | A fast and fully-automated deep-learning approach for accurate hemorrhage segmentation and volume quantification in non-contrast whole-head CT |
title_full | A fast and fully-automated deep-learning approach for accurate hemorrhage segmentation and volume quantification in non-contrast whole-head CT |
title_fullStr | A fast and fully-automated deep-learning approach for accurate hemorrhage segmentation and volume quantification in non-contrast whole-head CT |
title_full_unstemmed | A fast and fully-automated deep-learning approach for accurate hemorrhage segmentation and volume quantification in non-contrast whole-head CT |
title_short | A fast and fully-automated deep-learning approach for accurate hemorrhage segmentation and volume quantification in non-contrast whole-head CT |
title_sort | fast and fully-automated deep-learning approach for accurate hemorrhage segmentation and volume quantification in non-contrast whole-head ct |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7652921/ https://www.ncbi.nlm.nih.gov/pubmed/33168895 http://dx.doi.org/10.1038/s41598-020-76459-7 |
work_keys_str_mv | AT arabali afastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT chindabetty afastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT medvedevgeorge afastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT siuwilliam afastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT guohui afastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT gutao afastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT morenosylvain afastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT hamarnehghassan afastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT estermartin afastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT songxiaowei afastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT arabali fastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT chindabetty fastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT medvedevgeorge fastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT siuwilliam fastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT guohui fastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT gutao fastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT morenosylvain fastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT hamarnehghassan fastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT estermartin fastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct AT songxiaowei fastandfullyautomateddeeplearningapproachforaccuratehemorrhagesegmentationandvolumequantificationinnoncontrastwholeheadct |