Cargando…

A Topological Loss Function for Deep-Learning Based Image Segmentation Using Persistent Homology

We introduce a method for training neural networks to perform image or volume segmentation in which prior knowledge about the topology of the segmented object can be explicitly provided and then incorporated into the training process. By using the differentiable properties of persistent homology, a...

Descripción completa

Detalles Bibliográficos
Formato: Online Artículo Texto
Lenguaje:English
Publicado: IEEE 2020
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9721526/
https://www.ncbi.nlm.nih.gov/pubmed/32886606
http://dx.doi.org/10.1109/TPAMI.2020.3013679
_version_ 1784843796742144000
collection PubMed
description We introduce a method for training neural networks to perform image or volume segmentation in which prior knowledge about the topology of the segmented object can be explicitly provided and then incorporated into the training process. By using the differentiable properties of persistent homology, a concept used in topological data analysis, we can specify the desired topology of segmented objects in terms of their Betti numbers and then drive the proposed segmentations to contain the specified topological features. Importantly this process does not require any ground-truth labels, just prior knowledge of the topology of the structure being segmented. We demonstrate our approach in four experiments: one on MNIST image denoising and digit recognition, one on left ventricular myocardium segmentation from magnetic resonance imaging data from the UK Biobank, one on the ACDC public challenge dataset and one on placenta segmentation from 3-D ultrasound. We find that embedding explicit prior knowledge in neural network segmentation tasks is most beneficial when the segmentation task is especially challenging and that it can be used in either a semi-supervised or post-processing context to extract a useful training gradient from images without pixelwise labels.
format Online
Article
Text
id pubmed-9721526
institution National Center for Biotechnology Information
language English
publishDate 2020
publisher IEEE
record_format MEDLINE/PubMed
spelling pubmed-97215262022-12-06 A Topological Loss Function for Deep-Learning Based Image Segmentation Using Persistent Homology IEEE Trans Pattern Anal Mach Intell Article We introduce a method for training neural networks to perform image or volume segmentation in which prior knowledge about the topology of the segmented object can be explicitly provided and then incorporated into the training process. By using the differentiable properties of persistent homology, a concept used in topological data analysis, we can specify the desired topology of segmented objects in terms of their Betti numbers and then drive the proposed segmentations to contain the specified topological features. Importantly this process does not require any ground-truth labels, just prior knowledge of the topology of the structure being segmented. We demonstrate our approach in four experiments: one on MNIST image denoising and digit recognition, one on left ventricular myocardium segmentation from magnetic resonance imaging data from the UK Biobank, one on the ACDC public challenge dataset and one on placenta segmentation from 3-D ultrasound. We find that embedding explicit prior knowledge in neural network segmentation tasks is most beneficial when the segmentation task is especially challenging and that it can be used in either a semi-supervised or post-processing context to extract a useful training gradient from images without pixelwise labels. IEEE 2020-09-04 /pmc/articles/PMC9721526/ /pubmed/32886606 http://dx.doi.org/10.1109/TPAMI.2020.3013679 Text en https://creativecommons.org/licenses/by/4.0/This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
spellingShingle Article
A Topological Loss Function for Deep-Learning Based Image Segmentation Using Persistent Homology
title A Topological Loss Function for Deep-Learning Based Image Segmentation Using Persistent Homology
title_full A Topological Loss Function for Deep-Learning Based Image Segmentation Using Persistent Homology
title_fullStr A Topological Loss Function for Deep-Learning Based Image Segmentation Using Persistent Homology
title_full_unstemmed A Topological Loss Function for Deep-Learning Based Image Segmentation Using Persistent Homology
title_short A Topological Loss Function for Deep-Learning Based Image Segmentation Using Persistent Homology
title_sort topological loss function for deep-learning based image segmentation using persistent homology
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9721526/
https://www.ncbi.nlm.nih.gov/pubmed/32886606
http://dx.doi.org/10.1109/TPAMI.2020.3013679
work_keys_str_mv AT atopologicallossfunctionfordeeplearningbasedimagesegmentationusingpersistenthomology
AT atopologicallossfunctionfordeeplearningbasedimagesegmentationusingpersistenthomology
AT atopologicallossfunctionfordeeplearningbasedimagesegmentationusingpersistenthomology
AT atopologicallossfunctionfordeeplearningbasedimagesegmentationusingpersistenthomology
AT atopologicallossfunctionfordeeplearningbasedimagesegmentationusingpersistenthomology
AT atopologicallossfunctionfordeeplearningbasedimagesegmentationusingpersistenthomology
AT topologicallossfunctionfordeeplearningbasedimagesegmentationusingpersistenthomology
AT topologicallossfunctionfordeeplearningbasedimagesegmentationusingpersistenthomology
AT topologicallossfunctionfordeeplearningbasedimagesegmentationusingpersistenthomology
AT topologicallossfunctionfordeeplearningbasedimagesegmentationusingpersistenthomology
AT topologicallossfunctionfordeeplearningbasedimagesegmentationusingpersistenthomology
AT topologicallossfunctionfordeeplearningbasedimagesegmentationusingpersistenthomology