Cargando…

U-Net Architecture for Prostate Segmentation: The Impact of Loss Function on System Performance

Segmentation of the prostate gland from magnetic resonance images is rapidly becoming a standard of care in prostate cancer radiotherapy treatment planning. Automating this process has the potential to improve accuracy and efficiency. However, the performance and accuracy of deep learning models var...

Descripción completa

Detalles Bibliográficos
Autores principales: Montazerolghaem, Maryam, Sun, Yu, Sasso, Giuseppe, Haworth, Annette
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10135670/
https://www.ncbi.nlm.nih.gov/pubmed/37106600
http://dx.doi.org/10.3390/bioengineering10040412
_version_ 1785032033155678208
author Montazerolghaem, Maryam
Sun, Yu
Sasso, Giuseppe
Haworth, Annette
author_facet Montazerolghaem, Maryam
Sun, Yu
Sasso, Giuseppe
Haworth, Annette
author_sort Montazerolghaem, Maryam
collection PubMed
description Segmentation of the prostate gland from magnetic resonance images is rapidly becoming a standard of care in prostate cancer radiotherapy treatment planning. Automating this process has the potential to improve accuracy and efficiency. However, the performance and accuracy of deep learning models varies depending on the design and optimal tuning of the hyper-parameters. In this study, we examine the effect of loss functions on the performance of deep-learning-based prostate segmentation models. A U-Net model for prostate segmentation using T2-weighted images from a local dataset was trained and performance compared when using nine different loss functions, including: Binary Cross-Entropy (BCE), Intersection over Union (IoU), Dice, BCE and Dice (BCE + Dice), weighted BCE and Dice (W (BCE + Dice)), Focal, Tversky, Focal Tversky, and Surface loss functions. Model outputs were compared using several metrics on a five-fold cross-validation set. Ranking of model performance was found to be dependent on the metric used to measure performance, but in general, W (BCE + Dice) and Focal Tversky performed well for all metrics (whole gland Dice similarity coefficient (DSC): 0.71 and 0.74; 95HD: 6.66 and 7.42; Ravid 0.05 and 0.18, respectively) and Surface loss generally ranked lowest (DSC: 0.40; 95HD: 13.64; Ravid −0.09). When comparing the performance of the models for the mid-gland, apex, and base parts of the prostate gland, the models’ performance was lower for the apex and base compared to the mid-gland. In conclusion, we have demonstrated that the performance of a deep learning model for prostate segmentation can be affected by choice of loss function. For prostate segmentation, it would appear that compound loss functions generally outperform singles loss functions such as Surface loss.
format Online
Article
Text
id pubmed-10135670
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-101356702023-04-28 U-Net Architecture for Prostate Segmentation: The Impact of Loss Function on System Performance Montazerolghaem, Maryam Sun, Yu Sasso, Giuseppe Haworth, Annette Bioengineering (Basel) Article Segmentation of the prostate gland from magnetic resonance images is rapidly becoming a standard of care in prostate cancer radiotherapy treatment planning. Automating this process has the potential to improve accuracy and efficiency. However, the performance and accuracy of deep learning models varies depending on the design and optimal tuning of the hyper-parameters. In this study, we examine the effect of loss functions on the performance of deep-learning-based prostate segmentation models. A U-Net model for prostate segmentation using T2-weighted images from a local dataset was trained and performance compared when using nine different loss functions, including: Binary Cross-Entropy (BCE), Intersection over Union (IoU), Dice, BCE and Dice (BCE + Dice), weighted BCE and Dice (W (BCE + Dice)), Focal, Tversky, Focal Tversky, and Surface loss functions. Model outputs were compared using several metrics on a five-fold cross-validation set. Ranking of model performance was found to be dependent on the metric used to measure performance, but in general, W (BCE + Dice) and Focal Tversky performed well for all metrics (whole gland Dice similarity coefficient (DSC): 0.71 and 0.74; 95HD: 6.66 and 7.42; Ravid 0.05 and 0.18, respectively) and Surface loss generally ranked lowest (DSC: 0.40; 95HD: 13.64; Ravid −0.09). When comparing the performance of the models for the mid-gland, apex, and base parts of the prostate gland, the models’ performance was lower for the apex and base compared to the mid-gland. In conclusion, we have demonstrated that the performance of a deep learning model for prostate segmentation can be affected by choice of loss function. For prostate segmentation, it would appear that compound loss functions generally outperform singles loss functions such as Surface loss. MDPI 2023-03-26 /pmc/articles/PMC10135670/ /pubmed/37106600 http://dx.doi.org/10.3390/bioengineering10040412 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Montazerolghaem, Maryam
Sun, Yu
Sasso, Giuseppe
Haworth, Annette
U-Net Architecture for Prostate Segmentation: The Impact of Loss Function on System Performance
title U-Net Architecture for Prostate Segmentation: The Impact of Loss Function on System Performance
title_full U-Net Architecture for Prostate Segmentation: The Impact of Loss Function on System Performance
title_fullStr U-Net Architecture for Prostate Segmentation: The Impact of Loss Function on System Performance
title_full_unstemmed U-Net Architecture for Prostate Segmentation: The Impact of Loss Function on System Performance
title_short U-Net Architecture for Prostate Segmentation: The Impact of Loss Function on System Performance
title_sort u-net architecture for prostate segmentation: the impact of loss function on system performance
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10135670/
https://www.ncbi.nlm.nih.gov/pubmed/37106600
http://dx.doi.org/10.3390/bioengineering10040412
work_keys_str_mv AT montazerolghaemmaryam unetarchitectureforprostatesegmentationtheimpactoflossfunctiononsystemperformance
AT sunyu unetarchitectureforprostatesegmentationtheimpactoflossfunctiononsystemperformance
AT sassogiuseppe unetarchitectureforprostatesegmentationtheimpactoflossfunctiononsystemperformance
AT haworthannette unetarchitectureforprostatesegmentationtheimpactoflossfunctiononsystemperformance