Microwave seminar | 19 June 2023

 
Dr. David Bendahan
Aix Marseille University
Follow-up of patients with skeletal muscle disorders : How quantitative MRI and AI can be of any help?
Abstract

Due to its non invasive nature and its capacity to provide a good contrast between muscle and fat, MRI has been extensively used in the field of neuromuscular disorders and so for diagnostic and prognostic purposes. More recently quantitative approaches have been developed in order to compute outcome measures with a better sensitivity and specificity. Although these quantitative images are of large interest, the corresponding information makes sense if it can be computed from specific regions of interest which correspond to individual muscles of muscles group. This topographic information can only be assessed with dedicated segmentation methods. Manual segmentation is tedious, time-consuming and prone to inter-operator variability. If one intends to take advantage of the 3D nature of the MR images, automatic or semi-automatic segmentation methods should be developed. In this talk, these methods will be presented and the corresponding advantages and disadvantages will be described in the field of neuromuscular disorders.

1. Overview of MR Image Segmentation Strategies in Neuromuscular Disorders.
Ogier AC, Hostin MA, Bellemare ME, Bendahan D.
Front Neurol. 2021 Mar 25;12:625308. doi: 10.3389/fneur.2021.625308. eCollection 2021.
PMID: 33841299 Free PMC article. Review.
2. Medical image segmentation automatic quality control: A multi-dimensional approach.
Fournel J, Bartoli A, Bendahan D, Guye M, Bernard M, Rauseo E, Khanji MY, Petersen SE, Jacquier A, Ghattas B.
Med Image Anal. 2021 Dec;74:102213. doi: 10.1016/j.media.2021.102213. Epub 2021 Aug 12.
PMID: 34455223
3. Supervised segmentation framework for evaluation of diffusion tensor imaging indices in skeletal muscle.
Secondulfo L, Ogier AC, Monte JR, Aengevaeren VL, Bendahan D, Nederveen AJ, Strijkers GJ, Hooijmans MT.
NMR Biomed. 2021 Jan;34(1):e4406. doi: 10.1002/nbm.4406. Epub 2020 Oct 1.
PMID: 33001508 Free PMC article.
4. Quantification of Intra-Muscular Adipose Infiltration in Calf/Thigh MRI Using Fully and Weakly Supervised Semantic Segmentation.
Amer R, Nassar J, Trabelsi A, Bendahan D, Greenspan H, Ben-Eliezer N.
Bioengineering (Basel). 2022 Jul 14;9(7):315. doi: 10.3390/bioengineering9070315.
PMID: 35877366 Free PMC article.
5. The Impact of Fatty Infiltration on MRI Segmentation of Lower Limb Muscles in Neuromuscular Diseases: A Comparative Study of Deep Learning Approaches.
Hostin MA, Ogier AC, Michel CP, Le Fur Y, Guye M, Attarian S, Fortanier E, Bellemare ME, Bendahan D.
J Magn Reson Imaging. 2023 Apr 6. doi: 10.1002/jmri.28708. Online ahead of print.
PMID: 37025028