Related Experiment Video
Updated: Aug 5, 2026
09:30
Quantitative Magnetic Resonance Imaging of Skeletal Muscle Disease
Published on: December 18, 2016
3D Radiomic Texture Analysis of Quantitative Muscle MRI Enhances the Distinction Between Myotonic Dystrophy Type 1
Louise Iterbeke1, Lotte Huysmans2,3, Kobe Bamps4,5
1Laboratory for Muscle Diseases and Neuropathies, Department of Neurosciences, KU Leuven, and Leuven Brain Institute (LBI) and Leuven Institute for Rare Diseases (Leuven.IRD), Leuven, Belgium.
European Journal of Neurology
|August 1, 2026
Summary
Quantitative MRI texture analysis differentiates muscle diseases. This method reveals distinct fat infiltration patterns in myotonic dystrophy type 1 (DM1) and Charcot-Marie-Tooth neuropathy type 1A (CMT1A), improving diagnostic accuracy.
Area of Science:
- Neurology
- Radiology
- Medical Imaging
Background:
- Differentiating myogenic and neurogenic neuromuscular diseases (NMDs) presents clinical challenges.
- Quantitative muscle MRI (qMRI) using proton density fat fraction (PDFF) quantifies fat replacement but misses micro-spatial pathological patterns.
- This study explores 3D radiomic texture analysis (TA) with qMRI to improve NMD differentiation.
Purpose of the Study:
- To investigate if qMRI combined with 3D radiomic texture analysis can enhance the differentiation between myogenic and neurogenic diseases.
- To use myotonic dystrophy type 1 (DM1) and Charcot-Marie-Tooth neuropathy type 1A (CMT1A) as models for myogenic and neurogenic NMDs, respectively.
Main Methods:
- Thirty-three patients with DM1, 33 with CMT1A, and 33 healthy controls underwent qMRI using a 3T Philips Achieva system and a 6-point Dixon sequence.
- 3D segmentation of 28 lower limb muscles was performed using a convolutional neural network.
- Macroscopic features (volume, asymmetry, gradients) and micro-spatial radiomic features (entropy, contrast, homogeneity) were extracted from PDFF maps.
Main Results:
- Both DM1 and CMT1A cohorts showed significantly higher PDFF in lower limb muscles compared to controls.
- DM1 primarily affected the posterior compartment, while CMT1A affected the anterolateral compartment with steeper disto-proximal fat gradients.
- Radiomic TA revealed higher entropy and contrast, and lower homogeneity in CMT1A versus DM1, indicating distinct fat infiltration patterns (reticular in CMT1A, confluent in DM1).
Conclusions:
- 3D radiomic texture analysis of PDFF maps successfully identified distinct spatial fat replacement patterns in DM1 and CMT1A.
- Integrating radiomic features with conventional qMRI shows promise for non-invasive differentiation of DM1 and CMT1A.
- Further investigation in a wider range of NMDs is warranted.
