Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Apr:2021:1971-1974.
doi: 10.1109/isbi48211.2021.9434127. Epub 2021 May 25.

JOINT SEGMENTATION OF MULTIPLE SCLEROSIS LESIONS AND BRAIN ANATOMY IN MRI SCANS OF ANY CONTRAST AND RESOLUTION WITH CNNs

Benjamin Billot  1 , Stefano Cerri  2 , Koen Van Leemput  2   3 , Adrian V Dalca  3   4 , Juan Eugenio Iglesias  1   3   4
Affiliations

JOINT SEGMENTATION OF MULTIPLE SCLEROSIS LESIONS AND BRAIN ANATOMY IN MRI SCANS OF ANY CONTRAST AND RESOLUTION WITH CNNs

Benjamin Billot et al. Proc IEEE Int Symp Biomed Imaging. 2021 Apr.

Abstract

We present the first deep learning method to segment Multiple Sclerosis lesions and brain structures from MRI scans of any (possibly multimodal) contrast and resolution. Our method only requires segmentations to be trained (no images), as it leverages the generative model of Bayesian segmentation to generate synthetic scans with simulated lesions, which are then used to train a CNN. Our method can be retrained to segment at any resolution by adjusting the amount of synthesised partial volume. By construction, the synthetic scans are perfectly aligned with their labels, which enables training with noisy labels obtained with automatic methods. The training data are generated on the fly, and aggressive augmentation (including artefacts) is applied for improved generalisation. We demonstrate our method on two public datasets, comparing it with a state-of-the-art Bayesian approach implemented in FreeSurfer, and dataset specific CNNs trained on real data. The code is available at https://github.com/BBillot/SynthSeg.

Keywords: MS lesion; contrast-agnostic; segmentation.

PubMed Disclaimer

Figures

Fig. 1.
Fig. 1.
Generation of a FLAIR scan: (a) Training label map with MS lesions (bright purple); the segmentation of the brain regions are automated and thus imperfect. (b) Spatial augmentation. (c) GMM sampling, artefact and PV modelling.
Fig. 2.
Fig. 2.
(a) Box plots of the cross-validation Dice scores for the MS lesions and (c) for the average over 12 brain ROIs: cerebral cortex and white matter, lateral ventricle, cerebellar cortex and white matter, thalamus, caudate, putamen, pallidum, brainstem, hippocampus, and amygdala. (b) and (d) show the results obtained when training on MSSeg and testing on ISBI15.
Fig. 3.
Fig. 3.
Segmentation of an ISBI15 FLAIR scan: (a) ground truth, (b) supervised, (c) SAMSEG-lesion, (d) SynthSeg, (e) SynthSeg-rule, (f) SynthSeg-mix. MS lesions are in bright purple. Arrows indicate major segmentation errors (yellow for MS lesions, red for brain ROIs).
See this image and copyright information in PMC

References

    1. Confavreux C et al. “Rate of Pregnancy-Related Relapse in MS,” J. of Med, vol. 339, pp. 285–291, 1998. - PubMed
    1. Fisher E et al. “Gray matter atrophy in multiple sclerosis study,” Annals of Neur, vol. 64, pp. 255–265, 2008. - PubMed
    1. Barkhof F et al. “Imaging outcomes for neuroprotection and repair in multiple sclerosis trials,” Nature Reviews. Neurology, 5, pp. 256–266, 2009. - PubMed
    1. Commowick O et al. “Objective Evaluation of MS Lesion Segmentation using a Data Management and Processing Infrastructure,” Scientific Reports, vol. 8, 2018. - PMC - PubMed
    1. Kamnitsas K et al. “Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation,” Medical Im. Analysis, vol. 36, pp. 61–78, 2017. - PubMed

LinkOut - more resources