Options
Three-dimensional deep learning with spatial erasing for unsupervised anomaly segmentation in brain MRI
Citation Link: https://doi.org/10.15480/882.3761
Publikationstyp
Journal Article
Date Issued
2021-07-12
Sprache
English
TORE-DOI
Volume
16
Issue
9
Start Page
1413
End Page
1423
Citation
International Journal of Computer Assisted Radiology and Surgery 16 (9): 1413-1423 (2021-09)
Publisher DOI
Scopus ID
PubMed ID
34251654
Publisher
Springer
Purpose: Brain Magnetic Resonance Images (MRIs) are essential for the diagnosis of neurological diseases. Recently, deep learning methods for unsupervised anomaly detection (UAD) have been proposed for the analysis of brain MRI. These methods rely on healthy brain MRIs and eliminate the requirement of pixel-wise annotated data compared to supervised deep learning. While a wide range of methods for UAD have been proposed, these methods are mostly 2D and only learn from MRI slices, disregarding that brain lesions are inherently 3D and the spatial context of MRI volumes remains unexploited.
Methods: We investigate whether using increased spatial context by using MRI volumes combined with spatial erasing leads to improved unsupervised anomaly segmentation performance compared to learning from slices. We evaluate and compare 2D variational autoencoder (VAE) to their 3D counterpart, propose 3D input erasing, and systemically study the impact of the data set size on the performance. Results: Using two publicly available segmentation data sets for evaluation, 3D VAEs outperform their 2D counterpart, highlighting the advantage of volumetric context. Also, our 3D erasing methods allow for further performance improvements. Our best performing 3D VAE with input erasing leads to an average DICE score of 31.40% compared to 25.76% for the 2D VAE.
Conclusions: We propose 3D deep learning methods for UAD in brain MRI combined with 3D erasing and demonstrate that 3D methods clearly outperform their 2D counterpart for anomaly segmentation. Also, our spatial erasing method allows for further performance improvements and reduces the requirement for large data sets.
Methods: We investigate whether using increased spatial context by using MRI volumes combined with spatial erasing leads to improved unsupervised anomaly segmentation performance compared to learning from slices. We evaluate and compare 2D variational autoencoder (VAE) to their 3D counterpart, propose 3D input erasing, and systemically study the impact of the data set size on the performance. Results: Using two publicly available segmentation data sets for evaluation, 3D VAEs outperform their 2D counterpart, highlighting the advantage of volumetric context. Also, our 3D erasing methods allow for further performance improvements. Our best performing 3D VAE with input erasing leads to an average DICE score of 31.40% compared to 25.76% for the 2D VAE.
Conclusions: We propose 3D deep learning methods for UAD in brain MRI combined with 3D erasing and demonstrate that 3D methods clearly outperform their 2D counterpart for anomaly segmentation. Also, our spatial erasing method allows for further performance improvements and reduces the requirement for large data sets.
Subjects
3D autoencoder
Anomaly
Brain MRI
Segmentation
Unsupervised
MLE@TUHH
DDC Class
600: Technik
Funding(s)
More Funding Information
This work was partially funded by Grant Number ZF4026303TS9.
Publication version
publishedVersion
Loading...
Name
Bengs2021_Article_Three-dimensionalDeepLearningW.pdf
Size
3.45 MB
Format
Adobe PDF