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...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:IEEE transactions on pattern analysis and machine intelligence Ročník 44; číslo 12; s. 8766 - 8778
Hlavní autoři: Clough, James R., Byrne, Nicholas, Oksuz, Ilkay, Zimmer, Veronika A., Schnabel, Julia A., King, Andrew P.
Médium: Journal Article
Jazyk:angličtina
Vydáno: United States IEEE 01.12.2022
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:
Algorithms
convolutional neural networks
Data analysis
Deep Learning
Homology
Image Processing, Computer-Assisted - methods
Image segmentation
Labels
Loss measurement
Magnetic resonance imaging
Magnetic Resonance Imaging - methods
medical imaging
Myocardium
Network topologies
Network topology
Neural networks
Neural Networks, Computer
Object recognition
persistent homology
Segmentation
Shape
Topology
Training
ISSN:0162-8828, 1939-3539, 2160-9292, 1939-3539
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí: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.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:0162-8828
1939-3539
2160-9292
1939-3539
DOI:10.1109/TPAMI.2020.3013679