Deep 3D Convolutional Encoder Networks With Shortcuts for Multiscale Feature Integration Applied to Multiple Sclerosis Lesion Segmentation - PubMed (original) (raw)
. 2016 May;35(5):1229-1239.
doi: 10.1109/TMI.2016.2528821. Epub 2016 Feb 11.
- PMID: 26886978
- DOI: 10.1109/TMI.2016.2528821
Deep 3D Convolutional Encoder Networks With Shortcuts for Multiscale Feature Integration Applied to Multiple Sclerosis Lesion Segmentation
Tom Brosch et al. IEEE Trans Med Imaging. 2016 May.
Abstract
We propose a novel segmentation approach based on deep 3D convolutional encoder networks with shortcut connections and apply it to the segmentation of multiple sclerosis (MS) lesions in magnetic resonance images. Our model is a neural network that consists of two interconnected pathways, a convolutional pathway, which learns increasingly more abstract and higher-level image features, and a deconvolutional pathway, which predicts the final segmentation at the voxel level. The joint training of the feature extraction and prediction pathways allows for the automatic learning of features at different scales that are optimized for accuracy for any given combination of image types and segmentation task. In addition, shortcut connections between the two pathways allow high- and low-level features to be integrated, which enables the segmentation of lesions across a wide range of sizes. We have evaluated our method on two publicly available data sets (MICCAI 2008 and ISBI 2015 challenges) with the results showing that our method performs comparably to the top-ranked state-of-the-art methods, even when only relatively small data sets are available for training. In addition, we have compared our method with five freely available and widely used MS lesion segmentation methods (EMS, LST-LPA, LST-LGA, Lesion-TOADS, and SLS) on a large data set from an MS clinical trial. The results show that our method consistently outperforms these other methods across a wide range of lesion sizes.
Similar articles
- Improving automated multiple sclerosis lesion segmentation with a cascaded 3D convolutional neural network approach.
Valverde S, Cabezas M, Roura E, González-Villà S, Pareto D, Vilanova JC, Ramió-Torrentà L, Rovira À, Oliver A, Lladó X. Valverde S, et al. Neuroimage. 2017 Jul 15;155:159-168. doi: 10.1016/j.neuroimage.2017.04.034. Epub 2017 Apr 19. Neuroimage. 2017. PMID: 28435096 - Multi-branch convolutional neural network for multiple sclerosis lesion segmentation.
Aslani S, Dayan M, Storelli L, Filippi M, Murino V, Rocca MA, Sona D. Aslani S, et al. Neuroimage. 2019 Aug 1;196:1-15. doi: 10.1016/j.neuroimage.2019.03.068. Epub 2019 Apr 3. Neuroimage. 2019. PMID: 30953833 - Limited One-time Sampling Irregularity Map (LOTS-IM) for Automatic Unsupervised Assessment of White Matter Hyperintensities and Multiple Sclerosis Lesions in Structural Brain Magnetic Resonance Images.
Rachmadi MF, Valdés-Hernández MDC, Li H, Guerrero R, Meijboom R, Wiseman S, Waldman A, Zhang J, Rueckert D, Wardlaw J, Komura T. Rachmadi MF, et al. Comput Med Imaging Graph. 2020 Jan;79:101685. doi: 10.1016/j.compmedimag.2019.101685. Epub 2019 Nov 27. Comput Med Imaging Graph. 2020. PMID: 31846826 - VoxResNet: Deep voxelwise residual networks for brain segmentation from 3D MR images.
Chen H, Dou Q, Yu L, Qin J, Heng PA. Chen H, et al. Neuroimage. 2018 Apr 15;170:446-455. doi: 10.1016/j.neuroimage.2017.04.041. Epub 2017 Apr 23. Neuroimage. 2018. PMID: 28445774 Review. - Segmentation of human brain using structural MRI.
Helms G. Helms G. MAGMA. 2016 Apr;29(2):111-24. doi: 10.1007/s10334-015-0518-z. Epub 2016 Jan 6. MAGMA. 2016. PMID: 26739264 Review.
Cited by
- nnU-Net based segmentation and 3D reconstruction of uterine fibroids with MRI images for HIFU surgery planning.
Wang T, Wen Y, Wang Z. Wang T, et al. BMC Med Imaging. 2024 Sep 6;24(1):233. doi: 10.1186/s12880-024-01385-3. BMC Med Imaging. 2024. PMID: 39243001 Free PMC article. - Detection of diffusely abnormal white matter in multiple sclerosis on multiparametric brain MRI using semi-supervised deep learning.
Musall BC, Gabr RE, Yang Y, Kamali A, Lincoln JA, Jacobs MA, Ly V, Luo X, Wolinsky JS, Narayana PA, Hasan KM. Musall BC, et al. Sci Rep. 2024 Jul 26;14(1):17157. doi: 10.1038/s41598-024-67722-2. Sci Rep. 2024. PMID: 39060426 Free PMC article. - Deep learning to overcome Zernike phase-contrast nanoCT artifacts for automated micro-nano porosity segmentation in bone.
Silveira A, Greving I, Longo E, Scheel M, Weitkamp T, Fleck C, Shahar R, Zaslansky P. Silveira A, et al. J Synchrotron Radiat. 2024 Jan 1;31(Pt 1):136-149. doi: 10.1107/S1600577523009852. Epub 2024 Jan 1. J Synchrotron Radiat. 2024. PMID: 38095668 Free PMC article. - Investigation of an efficient multi-modal convolutional neural network for multiple sclerosis lesion detection.
Raab F, Malloni W, Wein S, Greenlee MW, Lang EW. Raab F, et al. Sci Rep. 2023 Nov 30;13(1):21154. doi: 10.1038/s41598-023-48578-4. Sci Rep. 2023. PMID: 38036638 Free PMC article. - 3D U-Net for automated detection of multiple sclerosis lesions: utility of transfer learning from other pathologies.
Wahlig SG, Nedelec P, Weiss DA, Rudie JD, Sugrue LP, Rauschecker AM. Wahlig SG, et al. Front Neurosci. 2023 Oct 27;17:1188336. doi: 10.3389/fnins.2023.1188336. eCollection 2023. Front Neurosci. 2023. PMID: 37965219 Free PMC article.
MeSH terms
LinkOut - more resources
Full Text Sources
Other Literature Sources
Medical
Miscellaneous