Cargando…
Strategies to improve deep learning-based salivary gland segmentation
BACKGROUND: Deep learning-based delineation of organs-at-risk for radiotherapy purposes has been investigated to reduce the time-intensiveness and inter-/intra-observer variability associated with manual delineation. We systematically evaluated ways to improve the performance and reliability of deep...
Autores principales: | , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
BioMed Central
2020
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7709305/ https://www.ncbi.nlm.nih.gov/pubmed/33261620 http://dx.doi.org/10.1186/s13014-020-01721-1 |
_version_ | 1783617721475268608 |
---|---|
author | van Rooij, Ward Dahele, Max Nijhuis, Hanne Slotman, Berend J. Verbakel, Wilko F. |
author_facet | van Rooij, Ward Dahele, Max Nijhuis, Hanne Slotman, Berend J. Verbakel, Wilko F. |
author_sort | van Rooij, Ward |
collection | PubMed |
description | BACKGROUND: Deep learning-based delineation of organs-at-risk for radiotherapy purposes has been investigated to reduce the time-intensiveness and inter-/intra-observer variability associated with manual delineation. We systematically evaluated ways to improve the performance and reliability of deep learning for organ-at-risk segmentation, with the salivary glands as the paradigm. Improving deep learning performance is clinically relevant with applications ranging from the initial contouring process, to on-line adaptive radiotherapy. METHODS: Various experiments were designed: increasing the amount of training data (1) with original images, (2) with traditional data augmentation and (3) with domain-specific data augmentation; (4) the influence of data quality was tested by comparing training/testing on clinical versus curated contours, (5) the effect of using several custom cost functions was explored, and (6) patient-specific Hounsfield unit windowing was applied during inference; lastly, (7) the effect of model ensembles was analyzed. Model performance was measured with geometric parameters and model reliability with those parameters’ variance. RESULTS: A positive effect was observed from increasing the (1) training set size, (2/3) data augmentation, (6) patient-specific Hounsfield unit windowing and (7) model ensembles. The effects of the strategies on performance diminished when the base model performance was already ‘high’. The effect of combining all beneficial strategies was an increase in average Sørensen–Dice coefficient of about 4% and 3% and a decrease in standard deviation of about 1% and 1% for the submandibular and parotid gland, respectively. CONCLUSIONS: A subset of the strategies that were investigated provided a positive effect on model performance and reliability. The clinical impact of such strategies would be an expected reduction in post-segmentation editing, which facilitates the adoption of deep learning for autonomous automated salivary gland segmentation. |
format | Online Article Text |
id | pubmed-7709305 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2020 |
publisher | BioMed Central |
record_format | MEDLINE/PubMed |
spelling | pubmed-77093052020-12-02 Strategies to improve deep learning-based salivary gland segmentation van Rooij, Ward Dahele, Max Nijhuis, Hanne Slotman, Berend J. Verbakel, Wilko F. Radiat Oncol Research BACKGROUND: Deep learning-based delineation of organs-at-risk for radiotherapy purposes has been investigated to reduce the time-intensiveness and inter-/intra-observer variability associated with manual delineation. We systematically evaluated ways to improve the performance and reliability of deep learning for organ-at-risk segmentation, with the salivary glands as the paradigm. Improving deep learning performance is clinically relevant with applications ranging from the initial contouring process, to on-line adaptive radiotherapy. METHODS: Various experiments were designed: increasing the amount of training data (1) with original images, (2) with traditional data augmentation and (3) with domain-specific data augmentation; (4) the influence of data quality was tested by comparing training/testing on clinical versus curated contours, (5) the effect of using several custom cost functions was explored, and (6) patient-specific Hounsfield unit windowing was applied during inference; lastly, (7) the effect of model ensembles was analyzed. Model performance was measured with geometric parameters and model reliability with those parameters’ variance. RESULTS: A positive effect was observed from increasing the (1) training set size, (2/3) data augmentation, (6) patient-specific Hounsfield unit windowing and (7) model ensembles. The effects of the strategies on performance diminished when the base model performance was already ‘high’. The effect of combining all beneficial strategies was an increase in average Sørensen–Dice coefficient of about 4% and 3% and a decrease in standard deviation of about 1% and 1% for the submandibular and parotid gland, respectively. CONCLUSIONS: A subset of the strategies that were investigated provided a positive effect on model performance and reliability. The clinical impact of such strategies would be an expected reduction in post-segmentation editing, which facilitates the adoption of deep learning for autonomous automated salivary gland segmentation. BioMed Central 2020-12-01 /pmc/articles/PMC7709305/ /pubmed/33261620 http://dx.doi.org/10.1186/s13014-020-01721-1 Text en © The Author(s) 2020 Open AccessThis 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/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. |
spellingShingle | Research van Rooij, Ward Dahele, Max Nijhuis, Hanne Slotman, Berend J. Verbakel, Wilko F. Strategies to improve deep learning-based salivary gland segmentation |
title | Strategies to improve deep learning-based salivary gland segmentation |
title_full | Strategies to improve deep learning-based salivary gland segmentation |
title_fullStr | Strategies to improve deep learning-based salivary gland segmentation |
title_full_unstemmed | Strategies to improve deep learning-based salivary gland segmentation |
title_short | Strategies to improve deep learning-based salivary gland segmentation |
title_sort | strategies to improve deep learning-based salivary gland segmentation |
topic | Research |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7709305/ https://www.ncbi.nlm.nih.gov/pubmed/33261620 http://dx.doi.org/10.1186/s13014-020-01721-1 |
work_keys_str_mv | AT vanrooijward strategiestoimprovedeeplearningbasedsalivaryglandsegmentation AT dahelemax strategiestoimprovedeeplearningbasedsalivaryglandsegmentation AT nijhuishanne strategiestoimprovedeeplearningbasedsalivaryglandsegmentation AT slotmanberendj strategiestoimprovedeeplearningbasedsalivaryglandsegmentation AT verbakelwilkof strategiestoimprovedeeplearningbasedsalivaryglandsegmentation |