PT  - JOURNAL ARTICLE
AU  - Avesta, A.
AU  - Hui, Y.
AU  - Aboian, M.
AU  - Duncan, J.
AU  - Krumholz, H.M.
AU  - Aneja, S.
TI  - 3D Capsule Networks for Brain Image Segmentation
AID  - 10.3174/ajnr.A7845
DP  - 2023 Apr 20
TA  - American Journal of Neuroradiology
4099  - http://www.ajnr.org/content/early/2023/04/20/ajnr.A7845.short
4100  - http://www.ajnr.org/content/early/2023/04/20/ajnr.A7845.full
AB  - BACKGROUND AND PURPOSE: Current autosegmentation models such as UNets and nnUNets have limitations, including the inability to segment images that are not represented during training and lack of computational efficiency. 3D capsule networks have the potential to address these limitations.MATERIALS AND METHODS: We used 3430 brain MRIs, acquired in a multi-institutional study, to train and validate our models. We compared our capsule network with standard alternatives, UNets and nnUNets, on the basis of segmentation efficacy (Dice scores), segmentation performance when the image is not well-represented in the training data, performance when the training data are limited, and computational efficiency including required memory and computational speed.RESULTS: The capsule network segmented the third ventricle, thalamus, and hippocampus with Dice scores of 95%, 94%, and 92%, respectively, which were within 1% of the Dice scores of UNets and nnUNets. The capsule network significantly outperformed UNets in segmenting images that were not well-represented in the training data, with Dice scores 30% higher. The computational memory required for the capsule network is less than one-tenth of the memory required for UNets or nnUNets. The capsule network is also >25% faster to train compared with UNet and nnUNet.CONCLUSIONS: We developed and validated a capsule network that is effective in segmenting brain images, can segment images that are not well-represented in the training data, and is computationally efficient compared with alternatives.CapsNetcapsule networkConv1first network layer made of convolutional operatorsConvCaps3third network layer made of convolutional capsulesConvCaps4fourth network layer made of convolutional capsulesDeconvCaps8eighth network layer made of deconvolutional capsulesFinalCaps13final thirteenth network layer made of capsulesFinalCaps13final layerGPUgraphics processing unitPrimaryCaps2second network layer made of primary capsules