Quantification of Intra-Muscular Adipose Infiltration in Calf/Thigh MRI Using Fully and Weakly Supervised Semantic Segmentation

Abstract

Infiltration of fat into lower limb muscles is one of the key markers for the severity of muscle pathologies. The level of fat infiltration varies in its severity across and within patients, and it is traditionally estimated using visual radiologic inspection. Precise quantification of the severity and spatial distribution of this pathological process requires accurate segmentation of lower limb anatomy into muscle and fat. Methods: Quantitative magnetic resonance imaging (qMRI) of the calf and thigh muscles is one of the most effective techniques for estimating pathological accumulation of intra-muscular adipose tissue (IMAT) in muscular dystrophies. In this work, we present a new deep learning (DL) network tool for automated and robust segmentation of lower limb anatomy that is based on the quantification of MRI’s transverse (T2) relaxation time. The network was used to segment calf and thigh anatomies into viable muscle areas and IMAT using a weakly supervised learning process. A new disease biomarker was calculated, reflecting the level of abnormal fat infiltration and disease state. A biomarker was then applied on two patient populations suffering from dysferlinopathy and Charcot–Marie–Tooth (CMT) diseases. Results: Comparison of manual vs. automated segmentation of muscle anatomy, viable muscle areas, and intermuscular adipose tissue (IMAT) produced high Dice similarity coefficients (DSCs) of 96.4%, 91.7%, and 93.3%, respectively. Linear regression between the biomarker value calculated based on the ground truth segmentation and based on automatic segmentation produced high correlation coefficients of 97.7% and 95.9% for the dysferlinopathy and CMT patients, respectively. Conclusions: Using a combination of qMRI and DL-based segmentation, we present a new quantitative biomarker of disease severity. This biomarker is automatically calculated and, most importantly, provides a spatially global indication for the state of the disease across the entire thigh or calf.

Decoding the transcriptome of muscular dystrophy due to Ptrf deficiency using single‐nucleus RNA sequencing

“…Adipocytes are the most common type of cells, apart from type IIb, in the muscle of Ptrf KO mice, and they are also one of the cell types with the most apparent changes in nuclei number between WT and Ptrf KO muscles (Figure 3E). Previous studies have shown adipocyte infiltration into muscular dystrophic muscles 22,66 . Thus, we next characterized adipocyte transcriptomic changes in response to muscular dystrophy.…”

“…Previous studies have shown adipocyte infiltration into muscular dystrophic muscles. 22,66 Thus, we next characterized adipocyte transcriptomic changes in response to muscular dystrophy. First, we reclustered adipocyte nuclei in Ptrf KO and WT muscle and found these nuclei were further divided into two subclusters (Figure 7A, left panel).…”

Estimation of subvoxel fat infiltration in neurodegenerative muscle disorders using quantitative multi‐T2 analysis

“…The FCNN's ground truth muscle and SC fat maps were delineated by an expert with more than 10 years of experience. The entire development process is described in Amer et al 26,54 A more rigorous development process where multiple experts provide the ground truth might have produced a more robust network. Previous work, however, used one expert for ground truth, while similar applications have been shown to perform well 54–58 …”

“…Previous work, however, used one expert for ground truth, while similar applications have been shown to perform well. [54][55][56][57][58] The fat component in the extended EMC algorithm was assumed to originate only from the methylene peak of the fat molecule. This is a limitation of this study because although this is the most abundant peak, the full fat spectrum contains other, smaller peaks.…”

AI in MRI: Computational Frameworks for a Faster, Optimized, and Automated Imaging Workflow

“…In this issue, Amer et al combined quantitative T 2 and proton density parameter maps with a multi-step classification pipeline aimed at segmenting and differentiating the various leg tissues [ 65 ]. By exploiting both fully and weakly supervised architectures, they were able to distinguish between the muscle, subcutaneous adipose, and infiltrated adipose tissues.…”