Texture as an imaging biomarker for disease severity in golden retriever muscular dystrophy

Abstract

Introduction: Golden retriever muscular dystrophy (GRMD), an X-linked recessive disorder, causes similar phenotypic features to Duchenne muscular dystrophy (DMD). There is currently a need for a quantitative and reproducible monitoring of disease progression for GRMD and DMD. Methods: To assess severity in the GRMD, we analyzed texture features extracted from multi-parametric MRI (T1w, T2w, T1m, T2m, and Dixon images) using 5 feature extraction methods and classified using support vector machines. Results: A single feature from qualitative images can provide 89% maximal accuracy. Furthermore, 2 features from T1w, T2m, or Dixon images provided highest accuracy. When considering a tradeoff between scan-time and computational complexity, T2m images provided good accuracy at a lower acquisition and processing time and effort. Conclusions: The combination of MRI texture features improved the classification accuracy for assessment of disease progression in GRMD with evaluation of the heterogenous nature of skeletal muscles as reflection of the histopathological changes.Muscle Nerve 59: [380][381][382][383][384][385][386] 2019 Duchenne muscular dystrophy (DMD) is an X-Additional supporting information may be found in the online version of this article.

Texture Analysis Of T1-Weighted Turbo Spin-Echo Mri For The Diagnosis And Follow-Up Of Collagen Vi-Related Myopathy

“…To overcome the lack of comprehensive human MRI data, many studies have focused on the Golden Retriever Muscular Dystrophy (GRMD) [8,9,10], which is considered to be the animal model with the highest similarity to Duchenne Muscular Dystrophy (DMD) observed in humans [6]. Duda et al [9] used GLCM-based features to detect GRMD at early stages (2-3 months old), achieving an accuracy above 0.90.…”

“…1) Gray-Level Co-occurrence Matrix: GLCM-based features [9,10] are among the most widely studied features for texture description. The GLCM encodes pairwise spatial relations between gray-level intensities, along four different directions (0 • , 45 • , 90 • and 135 • ) and considering distances up to 5 pixels.…”

“…2) Run-Length Matrix: The RLM [8,10] counts pixel runs along four different orientations -0 • , 45 • , 90 • and 135 • . P (g, l) refers to the number of runs with length l and gray level g. From the resulting P matrix of each orientation, the gray level non-uniformity, the run-length non-uniformity, the short run emphasis moment, the long run emphasis moment, and the fraction of image in runs were computed, yielding a total of 20 features.…”

Severity Classification in Cases of Collagen Vi-Related Myopathy with Convolutional Neural Networks and Handcrafted Texture Features

“…Texture features such as intensity histogram statistics, the Gray-Level Co-occurrence Matrix (GLCM), the Run Length Matrix (RLM), Local Binary Patterns, and Wavelet-based features have been used in a few studies [7,8,9,10,11] towards the development of CAD methods for muscle diseases using MRI, particularly the Duchenne Muscle Dystrophy (DMD) and the Golden Retriever Muscle Dystrophy (GRMD), which is considered to be highly similar to DMD [6].…”

Using MRI to quantify skeletal muscle pathology in Duchenne muscular dystrophy: A systematic mapping review

“…7 Due to its high soft-tissue contrast, high resolution, and absence of ionizing radiation, MRI has emerged as a promising noninvasive method for imaging skeletal muscles. 8 Various MRI sequences have been widely used to monitor DMD disease qualitatively 9,10 and quantitatively, [11][12][13][14] with a general hypothesis that structural changes in muscle will be reflected in MRI images. Qualitative analysis allows subjective grading of disease features, such as the level of signal intensity on T 1 weighted (T 1 w) and T 2 weighted (T 2 w) MRI protocols.…”