Accuracy of a machine learning muscle MRI-based tool for the diagnosis of muscular dystrophies

A first explainable-AI-based workflow integrating forward-forward and backpropagation-trained networks of label-free multiphoton microscopy images to assess human biopsies of rare neuromuscular disease

BACKGROUND AND OBJECTIVE

Diagnosis of rare neuromuscular diseases often relies on muscle biopsy analysis, which varies based on the evaluator's experience. Advances in deep learning show promise in improving diagnostic accuracy by identifying standardized features and phenotypic expressions in biopsy images. Explainable artificial intelligence extracts these features from the neural network's "black box," ensuring compliance with European ethical standards for the use of clinical data in real-world applications. This study proposes a clinic-friendly workflow, based on explainable artificial intelligence. It combines forward-forward and backpropagation-trained convolutional networks to identify complementary features of Duchenne Muscular Dystrophy. Our proposal sets the forward-forward training, applied here for the first time on biomedical images, as a potential new standard for interpretable deep learning applications in clinics.

METHODS

We analyzed a multiphoton microscopy dataset of 1600 images from 16 human muscle biopsies obtained during elective spinal surgery, combining autofluorescence, second and third harmonic generation signals. Class Activation Maps unveiled the dual decision-making process of the convolutional network, independently trained with both standard backpropagation and forward-forward algorithms. We evaluated the significance of the discovered features by using the Mann Whitney method. Entire biopsies were analyzed by providing an attention metric, computed as the weighted mean of all significant parameters.

RESULTS

Backpropagation, gold standard for 35 years, and forward-forward achieved over 90 % accuracy in distinguishing healthy and diseased patients tissue. Class activation maps revealed that, when trained independently with both algorithms, the same network identifies Duchenne Muscular Dystrophy tissue by focusing on different features. Both methods identified intramuscular collagen as a key feature. Backpropagation also highlighted collagen waviness and fatty tissue. Forward-forward emphasized collagen density. We integrated these complementary insights, validated by significance analysis, into a standardized attention metric, allowing a multi-structural, quantitative assessment of tissue changes and highlighting areas for further clinical analysis.

CONCLUSIONS

Our workflow, using clinically routine biopsies and transparent diagnostic features, demonstrates the unique potential of forward-forward in providing novel, reliable, interpretable results from biomedical images. This paves the way for its dual use with backpropagation, shifting benchmarks by enabling the discovery of features potentially overlooked by backpropagation across biomedical datasets.

Myo-Guide: A Machine Learning-Based Web Application for Neuromuscular Disease Diagnosis With MRI

BACKGROUND

Neuromuscular diseases (NMDs) are rare disorders characterized by progressive muscle fibre loss, leading to replacement by fibrotic and fatty tissue, muscle weakness and disability. Early diagnosis is critical for therapeutic decisions, care planning and genetic counselling. Muscle magnetic resonance imaging (MRI) has emerged as a valuable diagnostic tool by identifying characteristic patterns of muscle involvement. However, the increasing complexity of these patterns complicates their interpretation, limiting their clinical utility. Additionally, multi-study data aggregation introduces heterogeneity challenges. This study presents a novel multi-study harmonization pipeline for muscle MRI and an AI-driven diagnostic tool to assist clinicians in identifying disease-specific muscle involvement patterns.

METHODS

We developed a preprocessing pipeline to standardize MRI fat content across datasets, minimizing source bias. An ensemble of XGBoost models was trained to classify patients based on intramuscular fat replacement, age at MRI and sex. The SHapley Additive exPlanations (SHAP) framework was adapted to analyse model predictions and identify disease-specific muscle involvement patterns. To address class imbalance, training and evaluation were conducted using class-balanced metrics. The model's performance was compared against four expert clinicians using 14 previously unseen MRI scans.

RESULTS

Using our harmonization approach, we curated a dataset of 2961 MRI samples from genetically confirmed cases of 20 paediatric and adult NMDs. The model achieved a balanced accuracy of 64.8% ± 3.4%, with a weighted top-3 accuracy of 84.7% ± 1.8% and top-5 accuracy of 90.2% ± 2.4%. It also identified key features relevant for differential diagnosis, aiding clinical decision-making. Compared to four expert clinicians, the model obtained the highest top-3 accuracy (75.0% ± 4.8%). The diagnostic tool has been implemented as a free web platform, providing global access to the medical community.

CONCLUSIONS

The application of AI in muscle MRI for NMD diagnosis remains underexplored due to data scarcity. This study introduces a framework for dataset harmonization, enabling advanced computational techniques. Our findings demonstrate the potential of AI-based approaches to enhance differential diagnosis by identifying disease-specific muscle involvement patterns. The developed tool surpasses expert performance in diagnostic ranking and is accessible to clinicians worldwide via the Myo-Guide online platform.

MRI patterns of thigh muscle involvement in immune-mediated necrotizing myopathy and dermatomyositis

BACKGROUND

Immune-mediated necrotizing myopathy (IMNM) and dermatomyositis (DM) are characterized by weakness, hyperCKemia, associated autoantibodies, and varying extramuscular manifestations. Muscle MRI, currently subordinate to histopathology and serology in idiopathic inflammatory myopathy (IIM) classification, has an evolving role. Our study aims to define thigh muscle MRI involvement in IMNM and DM by direct comparison.

METHODS

This single-center, retrospective, cross-sectional study included 25 participants, who met IIM classification criteria (14 IMNM, 11 DM) and had available thigh MRI. Clinical and paraclinical data were available and reviewed. 11 muscles were graded for edema on MRI using a semi-quantitative scale (0: normal, 1: <30% of muscle involvement, 2: 31-75%, 3: > 75%). For 3 participants with no significant muscle edema, muscle fatty infiltration was scored according to the same scale. Using linear mixed-effects models, muscle scores were compared between the two groups and a secondary analysis was performed of only edema scores, excluding the 3 participants with fatty infiltration scores.

RESULTS

The most affected muscles in IMNM were the semimembranosus (3.0 [2.7-3.0] {median [IQR]}), biceps femoris-long head (BF-LH) (2.7 [2.0-3.0]), and adductors (2.5 [2.0-3.0]). In DM, the most affected muscles were the vastus lateralis (2.7 [2.3-3.0]), vastus intermedius (2.9 [2.2-3.0]), vastus medialis (2.3 [1.7-2.7]), semitendinosus (2.2 [1.0-2.7]), rectus femoris (RF) (2.0 [1.0-2.8]), biceps femoris-short head (BF-SH) (1.9 [1.0-2.7]), gracilis, and sartorius. Intergroup statistical difference of scores was significant (p < 0.01) for 10/11 thigh muscles excluding the RF (p = 0.19), supporting an inverse relationship of muscle involvement for DM and IMNM. The secondary comparative analysis of only muscle edema scores was significant (p < 0.05) for the same 10/11 muscles with a consistent direction for all comparisons.

CONCLUSION

DM and IMNM affect disparate thigh muscles on MRI. DM preferentially affects the anterior thigh, semitendinosus and BF-SH in the posterior thigh, and gracilis in the medial thigh, whereas IMNM preferentially affects the posterior thigh (semimembranosus and BF-LH) and adductors in the medial thigh.

A neuromuscular clinician's primer on machine learning

Artificial intelligence is the future of clinical practice and is increasingly utilized in medical management and clinical research. The release of ChatGPT3 in 2022 brought generative AI to the headlines and rekindled public interest in software agents that would complete repetitive tasks and save time. Artificial intelligence/machine learning underlies applications and devices which are assisting clinicians in the diagnosis, monitoring, formulation of prognosis, and treatment of patients with a spectrum of neuromuscular diseases. However, these applications have remained in the research sphere, and neurologists as a specialty are running the risk of falling behind other clinical specialties which are quicker to embrace these new technologies. While there are many comprehensive reviews on the use of artificial intelligence/machine learning in medicine, our aim is to provide a simple and practical primer to educate clinicians on the basics of machine learning. This will help clinicians specializing in neuromuscular and electrodiagnostic medicine to understand machine learning applications in nerve and muscle ultrasound, MRI imaging, electrical impendence myography, nerve conductions and electromyography and clinical cohort studies, and the limitations, pitfalls, regulatory and ethical concerns, and future directions. The question is not whether artificial intelligence/machine learning will change clinical practice, but when and how. How future neurologists will look back upon this period of transition will be determined not by how much changed or by how fast clinicians embraced this change but by how much patient outcomes were improved.

Automated analysis of quantitative muscle MRI and its reliability in patients with Duchenne muscular dystrophy

BACKGROUND

Quantitative muscle MRI is one of the most promising biomarkers to detect subclinical disease progression in patients with neuromuscular disorders, including Duchenne muscular dystrophy (DMD). However, its clinical application has been limited partly due to the time-intensive process of manual segmentation.

OBJECTIVE

We present a simple and fast automated approach to obtain quantitative measurement of thigh muscle fat fraction and investigate its reliability in patients with DMD.

METHODS

Clinical and radiological baseline and 6-month follow-up data of 41 ambulant patients with DMD were analysed retrospectively. Axial 2-point Dixon MR images of all thigh muscles were used to quantify mean fat fraction, while clinical outcomes were measured by the Motor Function Measure (MFM) and its D1 domain. Data obtained by automated segmentation were compared to manual segmentation and correlated with clinical outcomes. Results were also used to compare the statistical power when using automated or manual segmentation.

RESULTS

A mean increase of 3.55% in thigh muscle fat fraction at 6-month follow-up could be detected by both methods without any significant difference between them (p=0.437). The automated muscle segmentation method demonstrated a strong correlation with manually segmented data (Pearson's ρ = 0.97). Additionally, there was no statistically significant difference between the automated and manual segmentation methods in their association with clinical progression, as measured by the total MFM score and its D1 domain (p = 0.235 and p = 0.425, respectively).

CONCLUSIONS

The presented automated segmentation technique is a fast and reliable tool for assessing disease progression, particularly in the early stages of DMD. It is one of the few studies validated using manual segmentation, and with further refinement, it has the potential to become a good surrogate marker for disease progression in various neuromuscular disorders.

Whole-body muscle MRI in patients with spinal muscular atrophy

BACKGROUND

Spinal muscular atrophy (SMA) is a motor neuron disease with loss of musculature, which is replaced by fat. Previous magnetic resonance imaging (MRI) studies have focused on imaging muscles either in lower or upper extremities, but whole-body MRI can provide additional information on the involvement pattern. This study examined whole-body muscle fat replacement and the relationship between muscle structure, function, and bulbar symptoms.

METHOD

We conducted a descriptive, cross-sectional study. We assessed the fat replacement in skeletal muscles using whole-body MRI, the muscle function using the Motor Function Measurement 32, and bulbar muscle strength using the Bulbar Rating Scale. The presence of bulbar symptoms and function was assessed using the Voice Handicap Index, Eating Assessment Tool questionnaires, and a swallowing test.

RESULTS

We recruited 20 adult patients with type II and III SMA. The most affected muscles were the psoas major, soleus and rectus femoris, while the least affected muscles were the biceps brachii, deltoideus, and pterygoideus medialis. The tongue was involved in nearly half of the patients. Most patients reported issues with swallowing (75%) and voice (95%) but had relatively preserved strength of bulbar muscles.

CONCLUSION

Certain muscles are more prone to fat replacement than others in SMA, with a predominant proximal-distal and extensor-flexor involvement. Nearly half of the patients had increased fat content in the tongue, which is associated with dysphagia. In addition, most patients retained muscle strength in the bulbar muscles, despite advanced muscle weakness in the rest of the body.

Clinical and imaging spectrum of non-congenital dominant ACTN2 myopathy

BACKGROUND

Alpha-actinin-2, a protein with high expression in cardiac and skeletal muscle, is located in the Z-disc and plays a key role in sarcomere stability. Mutations in ACTN2 have been associated with both hypertrophic and dilated cardiomyopathy and, more recently, with skeletal myopathy.

METHODS

Genetic, clinical, and muscle imaging data were collected from 37 patients with an autosomal dominant ACTN2 myopathy belonging to 11 families from Spain and Belgium. Haplotypes were studied to confirm a common ancestry for the two most common recurrent variants identified in this study. A trained machine learning model was used to compare muscle MRI scans in ACTN2 myopathy with other neuromuscular diseases to identify a specific pattern of muscle involvement.

RESULTS

The clinical phenotype ranged from asymptomatic to limb-girdle weakness and facial involvement and was depending on genotype. Cardiac and respiratory involvement were not common. Belgian families carrying the p.Ile134Asn variant and Basque-Spanish families carrying the p.Cys487Arg variant each showed unique haplotypes supporting respective common ancestry. Available muscle biopsies showed non-specific changes. In muscle imaging, the most affected muscles were the glutei minor, glutei medius, hamstrings, tibialis anterior, and soleus. A machine learning model showed that the most differentiating features in dominant ACTN2 myopathy were the involvement of the tibialis anterior and gluteus medius muscles and preservation of the quadratus femoris, gastrocnemius lateralis, and tensor fasciae latae muscles.

CONCLUSION

We provide new insights into genetic, clinical, and muscle imaging characteristics of non-congenital dominant ACTN2 myopathy, broadening the phenotypic spectrum of muscle disorders caused by ACTN2 variants.

Report on the 6th Ottawa International Conference on Neuromuscular Disease & Biology - September 7-9, 2023, Ottawa, Canada

The 6th Ottawa International Conference in Neuromuscular Disease and Biology was held on September 7-9, 2023 in Ottawa, Canada. The goal of the conference was to assemble international experts in fundamental science, translational medicine and clinical neuromuscular disease research. Speakers provided attendees with updates on a wide range of topics related to neuromuscular disease and biology, including methods to identify novel diseases, recent developments in muscle, motor neuron and stem cell biology, expanded disease pathogenesis of known diseases, and exciting advances in therapy development. A summary of the major topics and results presented by these speakers is provided.

The diagnostic journey of patients being investigated for myopathy in a tertiary centre in England

Myopathies are heterogenous and can provide a diagnostic puzzle. Many patients investigated for myopathy will go on to other diagnoses. An overall understanding of how patients are investigated for suspected myopathy is lacking. Our aim was to understand how patients were investigated for myopathy in our tertiary centre and the timeline of their diagnostic journey. Through local database searches over a 5-year period (2015-2019), we identified a final total of 770 patients investigated for myopathy. Of these, 29.7% went on to a diagnosis of myopathy. The top non-myopathy diagnoses were neuropathy, spinal pathology and ataxia. Both the myopathy and non-myopathy groups had symptoms for an extended period before reaching specialist services (both groups 104 weeks). Following a first hospital visit, median time to diagnosis was not significantly different (myopathy 46.9 weeks, non-myopathy 40.7 weeks, p > 0.05). Data on the diagnostic journey for specific myopathies was also collected, with inflammatory myopathies diagnosed most quickly and muscular dystrophies most slowly. Muscle MRI and biopsy had the best positive predictive values (82.7% and 83.1%, respectively), while EMG had the best negative predictive value (89.3%). A combination of CK, EMG and neuroaxis MRI (brain and spinal cord) yielded at least one correct test result with respect to final diagnosis in 98.9% of cases. In conclusion, patients in whom a muscle disease is considered experience significant diagnostic delay. The first step in the diagnostic journey should be able to identify both myopathy and non-myopathy cases.

Myo-regressor Deep Informed Neural NetwOrk (Myo-DINO) for fast MR parameters mapping in neuromuscular disorders

Magnetic Resonance (MR) parameters mapping in muscle Magnetic Resonance Imaging (mMRI) is predominantly performed using pattern recognition-based algorithms, which are characterised by high computational costs and scalability issues in the context of multi-parametric mapping. Deep Learning (DL) has been demonstrated to be a robust and efficient method for rapid MR parameters mapping. However, its application in mMRI domain to investigate Neuromuscular Disorders (NMDs) has not yet been explored. In addition, data-driven DL models suffered in interpretation and explainability of the learning process. We developed a Physics Informed Neural Network called Myo-Regressor Deep Informed Neural NetwOrk (Myo-DINO) for efficient and explainable Fat Fraction (FF), water-T2 (wT2) and B1 mapping from a cohort of NMDs.A total of 2165 slices (232 subjects) from Multi-Echo Spin Echo (MESE) images were selected as the input dataset for which FF, wT2,B1 ground truth maps were computed using the MyoQMRI toolbox. This toolbox exploits the Extended Phase Graph (EPG) theory with a two-component model (water and fat signal) and slice profile to simulate the signal evolution in the MESE framework. A customized U-Net architecture was implemented as the Myo-DINO architecture. The squared L2 norm loss was complemented by two distinct physics models to define two 'Physics-Informed' loss functions: Cycling Loss 1 embedded a mono-exponential model to describe the relaxation of water protons, while Cycling Loss 2 incorporated the EPG theory with slice profile to model the magnetization dephasing under the effect of gradients and RF pulses. The Myo-DINO was trained with the hyperparameter value of the 'Physics-Informed' component held constant, i.e. λmodel = 1, while different hyperparameter values (λcnn) were applied to the squared L2 norm component in both the cycling loss. In particular, hard (λcnn=10), normal (λcnn=1) and self-supervised (λcnn=0) constraints were applied to gradually decrease the impact of the squared L2 norm component on the 'Physics Informed' term during the Myo-DINO training process. Myo-DINO achieved higher performance with Cycling Loss 2 for FF, wT2 and B1 prediction. In particular, high reconstruction similarity and quality (Structural Similarity Index > 0.92, Peak Signal to Noise ratio > 30.0 db) and small reconstruction error (Normalized Root Mean Squared Error < 0.038) to the reference maps were shown with self-supervised weighting of the Cycling Loss 2. In addition muscle-wise FF, wT2 and B1 predicted values showed good agreement with the reference values. The Myo-DINO has been demonstrated to be a robust and efficient workflow for MR parameters mapping in the context of mMRI. This provides preliminary evidence that it can be an effective alternative to the reference post-processing algorithm. In addition, our results demonstrate that Cycling Loss 2, which incorporates the Extended Phase Graph (EPG) model, provides the most robust and relevant physical constraints for Myo-DINO in this multi-parameter regression task. The use of Cycling Loss 2 with self-supervised constraint improved the explainability of the learning process because the network acquired domain knowledge solely in accordance with the assumptions of the EPG model.

MRI for the diagnosis of limb girdle muscular dystrophies

PURPOSE OF REVIEW

In the last 30 years, there have many publications describing the pattern of muscle involvement of different neuromuscular diseases leading to an increase in the information available for diagnosis. A high degree of expertise is needed to remember all the patterns described. Some attempts to use artificial intelligence or analysing muscle MRIs have been developed. We review the main patterns of involvement in limb girdle muscular dystrophies (LGMDs) and summarize the strategies for using artificial intelligence tools in this field.

RECENT FINDINGS

The most frequent LGMDs have a widely described pattern of muscle involvement; however, for those rarer diseases, there is still not too much information available. patients. Most of the articles still include only pelvic and lower limbs muscles, which provide an incomplete picture of the diseases. AI tools have efficiently demonstrated to predict diagnosis of a limited number of disease with high accuracy.

SUMMARY

Muscle MRI continues being a useful tool supporting the diagnosis of patients with LGMD and other neuromuscular diseases. However, the huge variety of patterns described makes their use in clinics a complicated task. Artificial intelligence tools are helping in that regard and there are already some accessible machine learning algorithms that can be used by the global medical community.

Real-time artificial intelligence-based texture analysis of muscle ultrasound data for neuromuscular disorder assessment

Objective

Many artificial intelligence approaches to muscle ultrasound image analysis have not been implemented on usable devices in clinical neuromuscular medicine practice, owing to high computational demands and lack of standardised testing protocols. This study evaluated the feasibility of using real-time texture analysis to differentiate between various pathological conditions.

Methods

We analysed 17,021 cross-sectional ultrasound images of the biceps brachii of 75 participants, including 25 each with neurogenic disorders, myogenic disorders, and healthy controls. The size and location of the regions of interest were randomly selected to minimise bias. A random forest classifier utilising texture features such as Dissimilarity and Homogeneity was developed and deployed on a mobile PC, enabling real-time analysis.

Results

The classifier distinguished patients with an accuracy of 81 %. Echogenicity and Contrast from the Co-Occurrence Matrix were significant predictive features. Validation on 15 patients achieved accuracies of 78 %/93 % per image/patient over 15-second videos, respectively. The use of a mobile PC facilitated real-time estimation of the underlying pathology during ultrasound examination, without influencing procedures.

Conclusions

Real-time automatic texture analysis is feasible as an adjunct for the diagnosis of neuromuscular disorders.

Significance

Artificial intelligence using texture analysis with a light computational load supports the semi-quantitative evaluation of neuromuscular ultrasound.

Conformational fingerprinting with Raman spectroscopy reveals protein structure as a translational biomarker of muscle pathology

Neuromuscular disorders are a group of conditions that can result in weakness of skeletal muscles. Examples include fatal diseases such as amyotrophic lateral sclerosis and conditions associated with high morbidity such as myopathies (muscle diseases). Many of these disorders are known to have abnormal protein folding and protein aggregates. Thus, easy to apply methods for the detection of such changes may prove useful diagnostic biomarkers. Raman spectroscopy has shown early promise in the detection of muscle pathology in neuromuscular disorders and is well suited to characterising the conformational profiles relating to protein secondary structure. In this work, we assess if Raman spectroscopy can detect differences in protein structure in muscle in the setting of neuromuscular disease. We utilise in vivo Raman spectroscopy measurements from preclinical models of amyotrophic lateral sclerosis and the myopathy Duchenne muscular dystrophy, together with ex vivo measurements of human muscle samples from individuals with and without myopathy. Using quantitative conformation profiling and matrix factorisation we demonstrate that quantitative 'conformational fingerprinting' can be used to identify changes in protein folding in muscle. Notably, myopathic conditions in both preclinical models and human samples manifested a significant reduction in α-helix structures, with concomitant increases in β-sheet and, to a lesser extent, nonregular configurations. Spectral patterns derived through non-negative matrix factorisation were able to identify myopathy with a high accuracy (79% in mouse, 78% in human tissue). This work demonstrates the potential of conformational fingerprinting as an interpretable biomarker for neuromuscular disorders.

Defining clinical endpoints in limb girdle muscular dystrophy: a GRASP-LGMD study

BACKGROUND

The Limb Girdle Muscular Dystrophies (LGMDs) are characterized by progressive weakness of the shoulder and hip girdle muscles as a result of over 30 different genetic mutations. This study is designed to develop clinical outcome assessments across the group of disorders.

METHODS/DESIGN

The primary goal of this study is to evaluate the utility of a set of outcome measures on a wide range of LGMD phenotypes and ability levels to determine if it would be possible to use similar outcomes between individuals with different phenotypes. We will perform a multi-center, 12-month study of 188 LGMD patients within the established Genetic Resolution and Assessments Solving Phenotypes in LGMD (GRASP-LGMD) Research Consortium, which is comprised of 11 sites in the United States and 2 sites in Europe. Enrolled patients will be clinically affected and have mutations in CAPN3 (LGMDR1), ANO5 (LGMDR12), DYSF (LGMDR2), DNAJB6 (LGMDD1), SGCA (LGMDR3), SGCB (LGMDR4), SGCD (LGMDR6), or SGCG (LGMDR5, or FKRP-related (LGMDR9).

DISCUSSION

To the best of our knowledge, this will be the largest consortium organized to prospectively validate clinical outcome assessments (COAs) in LGMD at its completion. These assessments will help clinical trial readiness by identifying reliable, valid, and responsive outcome measures as well as providing data driven clinical trial decision making for future clinical trials on therapeutic agents for LGMD. The results of this study will permit more efficient clinical trial design. All relevant data will be made available for investigators or companies involved in LGMD therapeutic development upon conclusion of this study as applicable.

TRIAL REGISTRATION

Clinicaltrials.gov NCT03981289; Date of registration: 6/10/2019.

Automatic segmentation of skeletal muscles from MR images using modified U-Net and a novel data augmentation approach

Rapid and accurate muscle segmentation is essential for the diagnosis and monitoring of many musculoskeletal diseases. As gold standard, manual annotation suffers from intensive labor and high inter-operator reproducibility errors. In this study, deep learning (DL) based automatic muscle segmentation from MR scans is investigated for post-menopausal women, who normally experience a decline in muscle volume. The performance of four Deep Learning (DL) models was evaluated: U-Net and UNet++ and two modified U-Net networks, which combined feature fusion and attention mechanisms (Feature-Fusion-UNet, FFU, and Attention-Feature-Fusion-UNet, AFFU). The models were tested for automatic segmentation of 16-lower limb muscles from MRI scans of two cohorts of post-menopausal women (11 subjects in PMW-1, 8 subjects in PMW-2; from two different studies so considered independent datasets) and 10 obese post-menopausal women (PMW-OB). Furthermore, a novel data augmentation approach is proposed to enlarge the training dataset. The results were assessed and compared by using the Dice similarity coefficient (DSC), relative volume error (RVE), and Hausdorff distance (HD). The best performance among all four DL models was achieved by AFFU (PMW-1: DSC 0.828 ± 0.079, 1-RVE 0.859 ± 0.122, HD 29.9 mm ± 26.5 mm; PMW-2: DSC 0.833 ± 0.065, 1-RVE 0.873 ± 0.105, HD 25.9 mm ± 27.9 mm; PMW-OB: DSC 0.862 ± 0.048, 1-RVE 0.919 ± 0.076, HD 34.8 mm ± 46.8 mm). Furthermore, the augmentation of data significantly improved the DSC scores of U-Net and AFFU for all 16 tested muscles (between 0.23% and 2.17% (DSC), 1.6%-1.93% (1-RVE), and 9.6%-19.8% (HD) improvement). These findings highlight the feasibility of utilizing DL models for automatic segmentation of muscles in post-menopausal women and indicate that the proposed augmentation method can enhance the performance of models trained on small datasets.

Magnetic resonance imaging-based criteria to differentiate dysferlinopathy from other genetic muscle diseases

The identification of disease-characteristic patterns of muscle fatty replacement in magnetic resonance imaging (MRI) is helpful for diagnosing neuromuscular diseases. In the Clinical Outcome Study of Dysferlinopathy, eight diagnostic rules were described based on MRI findings. Our aim is to confirm that they are useful to differentiate dysferlinopathy (DYSF) from other genetic muscle diseases (GMD). The rules were applied to 182 MRIs of dysferlinopathy patients and 1000 MRIs of patients with 10 other GMD. We calculated sensitivity (S), specificity (Sp), positive and negative predictive values (PPV/NPV) and accuracy (Ac) for each rule. Five of the rules were more frequently met by the DYSF group. Patterns observed in patients with FKRP, ANO5 and CAPN3 myopathies were similar to the DYSF pattern, whereas patterns observed in patients with OPMD, laminopathy and dystrophinopathy were clearly different. We built a model using the five criteria more frequently met by DYSF patients that obtained a S 95.9%, Sp 46.1%, Ac 66.8%, PPV 56% and NPV 94% to distinguish dysferlinopathies from other diseases. Our findings support the use of MRI in the diagnosis of dysferlinopathy, but also identify the need to externally validate "disease-specific" MRI-based diagnostic criteria using MRIs of other GMD patients.

Analysis of muscle magnetic resonance imaging of a large cohort of patient with VCP-mediated disease reveals characteristic features useful for diagnosis

BACKGROUND

The diagnosis of patients with mutations in the VCP gene can be complicated due to their broad phenotypic spectrum including myopathy, motor neuron disease and peripheral neuropathy. Muscle MRI guides the diagnosis in neuromuscular diseases (NMDs); however, comprehensive muscle MRI features for VCP patients have not been reported so far.

METHODS

We collected muscle MRIs of 80 of the 255 patients who participated in the "VCP International Study" and reviewed the T1-weighted (T1w) and short tau inversion recovery (STIR) sequences. We identified a series of potential diagnostic MRI based characteristics useful for the diagnosis of VCP disease and validated them in 1089 MRIs from patients with other genetically confirmed NMDs.

RESULTS

Fat replacement of at least one muscle was identified in all symptomatic patients. The most common finding was the existence of patchy areas of fat replacement. Although there was a wide variability of muscles affected, we observed a common pattern characterized by the involvement of periscapular, paraspinal, gluteal and quadriceps muscles. STIR signal was enhanced in 67% of the patients, either in the muscle itself or in the surrounding fascia. We identified 10 diagnostic characteristics based on the pattern identified that allowed us to distinguish VCP disease from other neuromuscular diseases with high accuracy.

CONCLUSIONS

Patients with mutations in the VCP gene had common features on muscle MRI that are helpful for diagnosis purposes, including the presence of patchy fat replacement and a prominent involvement of the periscapular, paraspinal, abdominal and thigh muscles.

Artificial Intelligence Applications in the Diagnosis of Neuromuscular Diseases: A Narrative Review

The accurate diagnosis of neuromuscular diseases (NMD) is in many cases difficult; the starting point is the clinical approach based on the course of the disease and a careful physical examination of the patient. Electrodiagnostic tests, imaging, muscle biopsy, and genetics are fundamental complementary studies for the diagnosis of NMD. The large volume of data obtained from such studies makes it necessary to look for efficient solutions, such as artificial intelligence (AI) applications, which can help classify, synthesize, and organize the information of patients with NMD to facilitate their accurate and timely diagnosis. The objective of this study was to describe the usefulness of AI applications in the diagnosis of patients with neuromuscular diseases. A narrative review was done, including publications on artificial intelligence applied to the diagnostic methods of NMD currently existing. Twelve studies were included. Two of the studies focused on muscle ultrasound, five of the studies on muscle MRI, two studies on electromyography, two studies on amyotrophic lateral sclerosis (ALS) biomarkers, and one study on genes related to myopathies. The accuracy of classification using different classification algorithms used in each of the studies included in this narrative review was already 90% in most studies. In conclusion, the future design of more accurate algorithms applied to NMD with greater precision will have an impact on the earlier diagnosis of this group of diseases.

Machine learning-based radiomics to differentiate immune-mediated necrotizing myopathy from limb-girdle muscular dystrophy R2 using MRI

Objectives

This study aimed to assess the feasibility of a machine learning-based radiomics tools to discriminate between Limb-girdle muscular dystrophy R2 (LGMDR2) and immune-mediated necrotizing myopathy (IMNM) using lower-limb muscle magnetic resonance imaging (MRI) examination.

Methods

After institutional review board approval, 30 patients with genetically proven LGMDR2 (12 females; age, 34.0 ± 11.3) and 45 patients with IMNM (28 females; age, 49.2 ± 16.6) who underwent lower-limb MRI examination including T1-weighted and interactive decomposition water and fat with echos asymmetric and least-squares estimation (IDEAL) sequences between July 2014 and August 2022 were included. Radiomics features of muscles were obtained, and four machine learning algorithms were conducted to select the optimal radiomics classifier for differential diagnosis. This selected algorithm was performed to construct the T1-weighted (TM), water-only (WM), or the combined model (CM) for calf-only, thigh-only, or the calf and thigh MR images, respectively. And their diagnostic performance was studied using area under the curve (AUC) and compared to the semi-quantitative model constructed by the modified Mercuri scale of calf and thigh muscles scored by two radiologists specialized in musculoskeletal imaging.

Results

The logistic regression (LR) model was the optimal radiomics model. The performance of the WM and CM for thigh-only images (AUC 0.893, 0.913) was better than those for calf-only images (AUC 0.846, 0.880) except the TM. For "calf + thigh" images, the TM, WM, and CM models always performed best (AUC 0.953, 0.907, 0.953) with excellent accuracy (92.0, 84.0, 88.0%). The AUCs of the Mercuri model of the calf, thigh, and "calf + thigh" images were 0.847, 0.900, and 0.953 with accuracy (84.0, 84.0, 88.0%).

Conclusion

Machine learning-based radiomics models can differentiate LGMDR2 from IMNM, performing better than visual assessment. The model built by combining calf and thigh images presents excellent diagnostic efficiency.

Deep learning based on susceptibility-weighted MR sequence for detecting cerebral microbleeds and classifying cerebral small vessel disease

BACKGROUND

Cerebral microbleeds (CMBs) serve as neuroimaging biomarkers to assess risk of intracerebral hemorrhage and diagnose cerebral small vessel disease (CSVD). Therefore, detecting CMBs can evaluate the risk of intracerebral hemorrhage and use its presence to support CSVD classification, both are conducive to optimizing CSVD management. This study aimed to develop and test a deep learning (DL) model based on susceptibility-weighted MR sequence (SWS) to detect CMBs and classify CSVD to assist neurologists in optimizing CSVD management. Patients with arteriolosclerosis (aSVD), cerebral amyloid angiopathy (CAA), and cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) treated at three centers were enrolled between January 2017 and May 2022 in this retrospective study. The SWSs of patients from two centers were used as the development set, and the SWSs of patients from the remaining center were used as the external test set. The DL model contains a Mask R-CNN for detecting CMBs and a multi-instance learning (MIL) network for classifying CSVD. The metrics for model performance included intersection over union (IoU), Dice score, recall, confusion matrices, receiver operating characteristic curve (ROC) analysis, accuracy, precision, and F1-score.

RESULTS

A total of 364 SWS were recruited, including 336 in the development set and 28 in the external test set. IoU for the model was 0.523 ± 0.319, Dice score 0.627 ± 0.296, and recall 0.706 ± 0.365 for CMBs detection in the external test set. For CSVD classification, the model achieved a weighted-average AUC of 0.908 (95% CI 0.895-0.921), accuracy of 0.819 (95% CI 0.768-0.870), weighted-average precision of 0.864 (95% CI 0.831-0.897), and weighted-average F1-score of 0.829 (95% CI 0.782-0.876) in the external set, outperforming the performance of the neurologist group.

CONCLUSION

The DL model based on SWS can detect CMBs and classify CSVD, thereby assisting neurologists in optimizing CSVD management.

Defining Clinical Endpoints in Limb Girdle Muscular Dystrophy: A GRASP-LGMD study

Background

The Limb Girdle Muscular Dystrophies (LGMDs) are characterized by progressive weakness of the shoulder and hip girdle muscles as a result of over 30 different genetic mutations. This study is designed to develop clinical outcome assessments across the group of disorders.

Methods/design

The primary goal of this study is to evaluate the utility of a set of outcome measures on a wide range of LGMD phenotypes and ability levels to determine if it would be possible to use similar outcomes between individuals with different phenotypes. We will perform a multi-center, 12-month study of 188 LGMD patients within the established Genetic Resolution and Assessments Solving Phenotypes in LGMD (GRASP-LGMD) Research Consortium, which is comprised of 11 sites in the United States and 2 sites in Europe. Enrolled patients will be clinically affected and have mutations in CAPN3 (LGMDR1), ANO5 (LGMDR12), DYSF (LGMDR2), DNAJB6 (LGMDD1), SGCA (LGMDR3), SGCB (LGMDR4), SGCD (LGMDR6), or SGCG (LGMDR5, or FKRP-related (LGMDR9).

Discussion

To the best of our knowledge, this will be the largest consortium organized to prospectively validate clinical outcome assessments (COAs) in LGMD at its completion. These assessments will help clinical trial readiness by identifying reliable, valid, and responsive outcome measures as well as providing data driven clinical trial decision making for future clinical trials on therapeutic agents for LGMD. The results of this study will permit more efficient clinical trial design. All relevant data will be made available for investigators or companies involved in LGMD therapeutic development upon conclusion of this study as applicable.

Trial registration

clinicaltrials.gov NCT03981289; Date of registration: 6/10/2019.

255th ENMC workshop: Muscle imaging in idiopathic inflammatory myopathies. 15th January, 16th January and 22nd January 2021 - virtual meeting and hybrid meeting on 9th and 19th September 2022 in Hoofddorp, The Netherlands

The 255th ENMC workshop on Muscle Imaging in Idiopathic Inflammatory myopathies (IIM) aimed at defining recommendations concerning the applicability of muscle imaging in IIM. The workshop comprised of clinicians, researchers and people living with myositis. We aimed to achieve consensus on the following topics: a standardized protocol for the evaluation of muscle images in various types of IIMs; the exact parameters, anatomical localizations and magnetic resonance imaging (MRI) techniques; ultrasound as assessment tool in IIM; assessment methods; the pattern of muscle involvement in IIM subtypes; the application of MRI as biomarker in follow-up studies and clinical trials, and the place of MRI in the evaluation of swallowing difficulty and cardiac manifestations. The following recommendations were formulated: In patients with suspected IIM, muscle imaging is highly recommended to be part of the initial diagnostic workup and baseline assessment. MRI is the preferred imaging modality due to its sensitivity to both oedema and fat accumulation. Ultrasound may be used for suspected IBM. Repeat imaging should be considered if patients do not respond to treatment, if there is ongoing diagnostic uncertainty or there is clinical or laboratory evidence of disease relapse. Quantitative MRI is established as a sensitive biomarker in IBM and could be included as a primary or secondary outcome measure in early phase clinical trials, or as a secondary outcome measure in late phase clinical trials. Finally, a research agenda was drawn up.

Advantages of digital technology in the assessment of bone marrow involvement in Gaucher's disease

Gaucher disease (GD) is a genetic lysosomal disorder characterized by high bone marrow (BM) involvement and skeletal complications. The pathophysiology of these complications is not fully elucidated. Magnetic resonance imaging (MRI) is the gold standard to evaluate BM. This study aimed to apply machine-learning techniques in a cohort of Spanish GD patients by a structured bone marrow MRI reporting model at diagnosis and follow-up to predict the evolution of the bone disease. In total, 441 digitalized MRI studies from 131 patients (M: 69, F:62) were reevaluated by a blinded expert radiologist who applied a structured report template. The studies were classified into categories carried out at different stages as follows: A: baseline; B: between 1 and 4 y of follow-up; C: between 5 and 9 y; and D: after 10 years of follow-up. Demographics, genetics, biomarkers, clinical data, and cumulative years of therapy were included in the model. At the baseline study, the mean age was 37.3 years (1-80), and the median Spanish MRI score (S-MRI) was 8.40 (male patients: 9.10 vs. female patients: 7.71) (p < 0.001). BM clearance was faster and deeper in women during follow-up. Genotypes that do not include the c.1226A>G variant have a higher degree of infiltration and complications (p = 0.017). A random forest machine-learning model identified that BM infiltration degree, age at the start of therapy, and femur infiltration were the most important factors to predict the risk and severity of the bone disease. In conclusion, a structured bone marrow MRI reporting in GD is useful to standardize the collected data and facilitate clinical management and academic collaboration. Artificial intelligence methods applied to these studies can help to predict bone disease complications.

Provisional practice recommendation for the management of myopathy in VCP-associated multisystem proteinopathy

Valosin-containing protein (VCP)-associated multisystem proteinopathy (MSP) is a rare genetic disorder with abnormalities in the autophagy pathway leading to various combinations of myopathy, bone diseases, and neurodegeneration. Ninety percent of patients with VCP-associated MSP have myopathy, but there is no consensus-based guideline. The goal of this working group was to develop a best practice set of provisional recommendations for VCP myopathy which can be easily implemented across the globe. As an initiative by Cure VCP Disease Inc., a patient advocacy organization, an online survey was initially conducted to identify the practice gaps in VCP myopathy. All prior published literature on VCP myopathy was reviewed to better understand the different aspects of management of VCP myopathy, and several working group sessions were conducted involving international experts to develop this provisional recommendation. VCP myopathy has a heterogeneous clinical phenotype and should be considered in patients with limb-girdle muscular dystrophy phenotype, or any myopathy with an autosomal dominant pattern of inheritance. Genetic testing is the only definitive way to diagnose VCP myopathy, and single-variant testing in the case of a known familial VCP variant, or multi-gene panel sequencing in undifferentiated cases can be considered. Muscle biopsy is important in cases of diagnostic uncertainty or lack of a definitive pathogenic genetic variant since rimmed vacuoles (present in ~40% cases) are considered a hallmark of VCP myopathy. Electrodiagnostic studies and magnetic resonance imaging can also help rule out disease mimics. Standardized management of VCP myopathy will optimize patient care and help future research initiatives.

Roles of Skeletal Muscle in Development: A Bioinformatics and Systems Biology Overview

The ability to assess various cellular events consequent to perturbations, such as genetic mutations, disease states and therapies, has been recently revolutionized by technological advances in multiple "omics" fields. The resulting deluge of information has enabled and necessitated the development of tools required to both process and interpret the data. While of tremendous value to basic researchers, the amount and complexity of the data has made it extremely difficult to manually draw inference and identify factors key to the study objectives. The challenges of data reduction and interpretation are being met by the development of increasingly complex tools that integrate disparate knowledge bases and synthesize coherent models based on current biological understanding. This chapter presents an example of how genomics data can be integrated with biological network analyses to gain further insight into the developmental consequences of genetic perturbations. State of the art methods for conducting similar studies are discussed along with modern methods used to analyze and interpret the data.

Artificial intelligence, machine learning and deep learning in musculoskeletal imaging: Current applications

Artificial intelligence is rapidly expanding in all technological fields. The medical field, and especially diagnostic imaging, has been showing the highest developmental potential. Artificial intelligence aims at human intelligence simulation through the management of complex problems. This review describes the technical background of artificial intelligence, machine learning, and deep learning. The first section illustrates the general potential of artificial intelligence applications in the context of request management, data acquisition, image reconstruction, archiving, and communication systems. In the second section, the prospective of dedicated tools for segmentation, lesion detection, automatic diagnosis, and classification of musculoskeletal disorders is discussed.

Neuromuscular disorders: finding the missing genetic diagnoses

Neuromuscular disorders (NMDs) are a wide-ranging group of diseases that seriously affect the quality of life of affected individuals. The development of next-generation sequencing revolutionized the diagnosis of NMD, enabling the discovery of hundreds of NMD genes and many more pathogenic variants. However, the diagnostic yield of genetic testing in NMD cohorts remains incomplete, indicating a large number of genetic diagnoses are not identified through current methods. Fortunately, recent advancements in sequencing technologies, analytical tools, and high-throughput functional screening provide an opportunity to circumvent current challenges. Here, we discuss reasons for missing genetic diagnoses in NMD, how emerging technologies and tools can overcome these hurdles, and examine future approaches to improving diagnostic yields in NMD.

Structured bone marrow report as an assessment tool in patients with hematopoietic disorders

BACKGROUND

There are multiple hematological and other entities (metastases, infections) that can affect the bone marrow (BM). The gold standard imaging technique for BM examination is magnetic resonance imaging (MRI). Technological advances have made it possible to digitalize image files and create applications that help to produce higher quality structured reports, facilitating the analysis of data and unifying the criteria collected, making it possible to fill an existing gap. The aim of this study is to present a structured report model applicable to BM studies by MRI.

METHODS

We have carried out a systematic review following the recommendations of the PRISMA checklist report to explore previous publications applying structured BM MRI reporting. Eligibility criteria: the selection of articles carried out by MeSH thesaurus. Original or review articles of BM pathology assessed by MRI. Our group with a wide experience in the evaluation of BM by MRI have designed a model for BM report using eight items: demographic data, diagnostic suspicion, technical data, type of exam initial or control, distribution and patterns involvement, complications and location, total assessment comments.

RESULTS

We have not found articles that reflect the existence of a structured report of BM examination by MRI. Only one descriptive article has been identified on guidelines for acquisition, interpretation and reporting which refers to a single entity. With the selected parameters, a software has been developed that allows to fill in the sections of the structured report with ease and immediacy and to send the result directly to the clinician.

DISCUSSION

Structured reports are the result of applying a logical structure to the radiological report, and the rules of elaboration comprise several criteria: (I) using a uniform language. The standardization of terminology avoids ambiguity in reporting and makes it easier to compare reports. (II) Accurately describe the radiological findings, following a prescribed order with review questions and answers. (III) Drafting using diagnostic screening tables. (IV) Respect the radiologists' workflow by facilitating the work and not hindering it. The final report of this work has been the product of the clinical-radiological collaboration in our working group.

Muscle MRI characteristic pattern for late-onset TK2 deficiency diagnosis

Background and objective

TK2 deficiency (TK2d) is a rare mitochondrial disorder that manifests predominantly as a progressive myopathy with a broad spectrum of severity and age of onset. The rate of progression is variable, and the prognosis is poor due to early and severe respiratory involvement. Early and accurate diagnosis is particularly important since a specific treatment is under development. This study aims to evaluate the diagnostic value of lower limb muscle MRI in adult patients with TK2d.

Methods

We studied a cohort of 45 genetically confirmed patients with mitochondrial myopathy (16 with mutations in TK2, 9 with mutations in other nuclear genes involved in mitochondrial DNA [mtDNA] synthesis or maintenance, 10 with single mtDNA deletions, and 10 with point mtDNA mutations) to analyze the imaging pattern of fat replacement in lower limb muscles. We compared the identified pattern in patients with TK2d with the MRI pattern of other non-mitochondrial genetic myopathies that share similar clinical characteristics.

Results

We found a consistent lower limb muscle MRI pattern in patients with TK2d characterized by involvement of the gluteus maximus, gastrocnemius medialis, and sartorius muscles. The identified pattern in TK2 patients differs from the known radiological involvement of other resembling muscle dystrophies that share clinical features.

Conclusions

By analyzing the largest cohort of muscle MRI from patients with mitochondrial myopathies studied to date, we identified a characteristic and specific radiological pattern of muscle involvement in patients with TK2d that could be useful to speed up its diagnosis.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00415-021-10957-0.

Deep convolutional neural network-based algorithm for muscle biopsy diagnosis

Histopathologic evaluation of muscle biopsy samples is essential for classifying and diagnosing muscle diseases. However, the numbers of experienced specialists and pathologists are limited. Although new technologies such as artificial intelligence are expected to improve medical reach, their use with rare diseases, such as muscle diseases, is challenging because of the limited availability of training datasets. To address this gap, we developed an algorithm based on deep convolutional neural networks (CNNs) and collected 4041 microscopic images of 1400 hematoxylin-and-eosin-stained pathology slides stored in the National Center of Neurology and Psychiatry for training CNNs. Our trained algorithm differentiated idiopathic inflammatory myopathies (mostly treatable) from hereditary muscle diseases (mostly non-treatable) with an area under the curve (AUC) of 0.996 and achieved better sensitivity and specificity than the diagnoses done by nine physicians under limited diseases and conditions. Furthermore, it successfully and accurately classified four subtypes of the idiopathic inflammatory myopathies with an average AUC of 0.958 and classified seven subtypes of hereditary muscle disease with an average AUC of 0.936. We also established a method to validate the similarity between the predictions made by the algorithm and the seven physicians using visualization technology and clarified the validity of the predictions. These results support the reliability of the algorithm and suggest that our algorithm has the potential to be used straightforwardly in a clinical setting.

Muscle MRI in Myotonic Dystrophy type 1 (DM1): refining muscle involvement and implications for clinical trials

BACKGROUND

Only few studies reported muscle imaging data on small cohorts of patients with Myotonic dystrophy type 1 (DM1). We aimed to investigate the muscle involvement in a large cohort of patients, to refine the pattern of muscle involvement, to better understand the pathophysiological mechanisms of muscle weakness and to identify potential imaging biomarkers for disease activity and severity.

METHODS

134 DM1 patients underwent a cross-sectional muscle MRI study. STIR and T1- sequences in lower and upper body were analysed. Fat replacement, muscle atrophy and STIR positivity were evaluated using three different scales. Correlations between MRI scores, clinical features and genetic background were investigated.

RESULTS

The most frequent pattern of muscle involvement in T1 consisted of fat replacement of the tongue, sternocleidomastoideus, paraspinalis, gluteus minimus, distal quadriceps and gastrocnemius medialis. Degree of fat replacement at MRI correlated with clinical severity and disease duration, but not with CTG expansion. Fat replacement was also detected in milder/asymptomatic patients. More than 80% of patients had STIR positive signal in muscles. Most DM1 patients also showed a variable degree of muscle atrophy regardless MRI signs of fat replacement. A subset of patients (20%) showed a "marbled" muscle appearance.

CONCLUSIONS

muscle MRI is a sensitive biomarker of disease severity also for the milder spectrum of disease. STIR hyperintensty seems to precede fat replacement in T1. Beyond fat replacement, STIR positivity, muscle atrophy and "marbled" appearance suggest further mechanisms of muscle wasting and weakness in DM1, representing additional outcome measures and therapeutical targets for forthcoming clinical trials.

Two decades of advances in muscle imaging in children: from pattern recognition of muscle diseases to quantification and machine learning approaches

Muscle imaging has progressively gained popularity in the neuromuscular field. Together with detailed clinical examination and muscle biopsy, it has become one of the main tools for deep phenotyping and orientation of etiological diagnosis. Even in the current era of powerful new generation sequencing, muscle MRI has arisen as a tool for prioritization of certain genetic entities, supporting the pathogenicity of variants of unknown significance and facilitating diagnosis in cases with an initially inconclusive genetic study. Although the utility of muscle imaging is increasingly clear, it has not reached its full potential in clinical practice. Pattern recognition is known for a number of diseases and will certainly be enhanced by the use of machine learning approaches. For instance, MRI heatmap representations might be confronted with molecular results by obtaining a probabilistic diagnosis based in each disease "MRI fingerprints". Muscle ultrasound as a screening tool and quantified techniques such as Dixon MRI seem still underdeveloped. In this paper, we aim to appraise the advances in recent years in pediatric muscle imaging and try to define areas of uncertainty and potential advances that might become standardized to be widely used in the future.

Muscle magnetic resonance imaging in patients with LAMA2-related muscular dystrophy

LAMA2-related muscular dystrophy (LAMA2-MD) is classified into congenital muscular dystrophy type 1A (MDC1A) and autosomal recessive limb-girdle muscular dystrophy-23 (LGMDR23). The purpose of this study was to identify the involvement pattern of thigh muscles of LAMA2-MD patients on magnetic resonance imaging. Fourteen MDC1A and 3 LGMDR23 patients were included, with 21 known and 8 novel LAMA2 disease-causing variants. In LAMA2-MD, the gluteus maximus, anterior (quadriceps femoris) and posterior (adductor magnus and biceps femoris) thigh muscles were extensively and severely affected with fatty infiltration, with relatively sparing of the adductor longus. The pattern of muscle involvement was similar between MDC1A and LGMDR23, but more severe in MDC1A, as well as in LAMA2-MD patients without ambulation. The rather peculiar pattern of the adductor magnus and long head of the biceps femoris first and severely affected in the mid-thigh level was found in LGMDR23. Strong correlation between fatty infiltration and age as well as disease duration was observed for the adductor longus in MDC1A. Edema and atrophy selectively involved in some muscles. The pattern of fatty infiltration on thigh muscle MRI of LAMA2-MD could provide important information for the diagnosis, differential diagnosis and assessment of clinical severity.

Diagnostic magnetic resonance imaging biomarkers for facioscapulohumeral muscular dystrophy identified by machine learning

BACKGROUND

The diagnosis of facioscapulohumeral muscular dystrophy (FSHD) can be challenging in patients not displaying the classical phenotype or with atypical clinical features. Despite the identification by magnetic resonance imaging (MRI) of selective patterns of muscle involvement, their specificity and added diagnostic value are unknown.

METHODS

We aimed to identify the radiological features more useful to distinguish FSHD from other myopathies and test the diagnostic accuracy of MRI. A retrospective cohort of 295 patients (187 FSHD, 108 non-FSHD) studied by upper and lower-limb muscle MRI was analyzed. Scans were evaluated for the presence of 15 radiological features. A random forest machine learning algorithm was used to identify the most relevant for FSHD diagnosis. Different patterns were created by their combination and diagnostic accuracy of each of them was tested.

RESULTS

The combination of trapezius involvement and bilateral subscapularis muscle sparing achieved the best diagnostic accuracy (0.89, 95% Confidence Interval [0.85-0.92]) with 0.90 [0.85-0.94] sensitivity and 0.88 [0.80-0.93] specificity. This pattern correctly identified 91% atypical FSHD patients of our cohort. The combination of trapezius involvement, bilateral subscapularis and iliopsoas sparing and asymmetric involvement of upper and lower-limb muscles was pathognomonic for FSHD, yielding a specificity of 0.99 [0.95-1.00].

CONCLUSIONS

We identified MRI patterns that showed a high diagnostic power in promptly discriminating FSHD from other muscle disorders, with comparable performance irrespective of typical or atypical clinical features. Upper girdle in addition to lower-limb muscle imaging should be extensively implemented in the diagnostic workup to support or exclude a diagnosis of FSHD.

Plasma lipidomic analysis shows a disease progression signature in mdx mice

Duchenne muscular dystrophy (DMD) is a rare genetic disorder affecting paediatric patients. The disease course is characterized by loss of muscle mass, which is rapidly substituted by fibrotic and adipose tissue. Clinical and preclinical models have clarified the processes leading to muscle damage and myofiber degeneration. Analysis of the fat component is however emerging as more evidence shows how muscle fat fraction is associated with patient performance and prognosis. In this article we aimed to study whether alterations exist in the composition of lipids in plasma samples obtained from mouse models. Analysis of plasma samples was performed in 4 mouse models of DMD and wild-type mice by LC–MS. Longitudinal samplings of individual mice covering an observational period of 7 months were obtained to cover the different phases of the disease. We report clear elevation of glycerolipids and glycerophospholipids families in dystrophic mice compared to healthy mice. Triacylglycerols were the strongest contributors to the signatures in mice. Annotation of individual lipids confirmed the elevation of lipids belonging to these families as strongest discriminants between healthy and dystrophic mice. A few sphingolipids (such as ganglioside GM2, sphingomyelin and ceramide), sterol lipids (such as cholesteryl oleate and cholesteryl arachidonate) and a fatty acyl (stearic acid) were also found to be affected in dystrophic mice. Analysis of serum and plasma samples show how several lipids are affected in dystrophic mice affected by muscular dystrophy. This study sets the basis to further investigations to understand how the lipid signature relates to the disease biology and muscle performance.

The increasing role of muscle MRI to monitor changes over time in untreated and treated muscle diseases

PURPOSE OF REVIEW

This review aims to discuss the recent results of studies published applying quantitative MRI sequences to large cohorts of patients with neuromuscular diseases.

RECENT FINDINGS

Quantitative MRI sequences are now available to identify and quantify changes in muscle water and fat content. These two components have been associated with acute and chronic injuries, respectively. Studies show that the increase in muscle water is not only reversible if therapies are applied successfully but can also predict fat replacement in neurodegenerative diseases. Muscle fat fraction correlates with muscle function tests and increases gradually over time in parallel with the functional decline of patients with neuromuscular diseases. There are new spectrometry-based sequences to quantify other components, such as glycogen, electrolytes or the pH of the muscle fibre, extending the applicability of MRI to the study of several processes in neuromuscular diseases.

SUMMARY

The latest results obtained from the study of long cohorts of patients with various neuromuscular diseases open the door to the use of this technology in clinical trials, which would make it possible to obtain a new measure for assessing the effectiveness of new treatments. The challenge is currently the popularization of these studies and their application to the monitoring of patients in the daily clinic.

MR imaging of inherited myopathies: a review and proposal of imaging algorithms

PURPOSE OF REVIEW

The aims of this review are to discuss the imaging modalities used to assess muscle changes in myopathies, to provide an overview of the inherited myopathies focusing on their patterns of muscle involvement in magnetic resonance imaging (MR), and to propose up-to-date imaging-based diagnostic algorithms that can help in the diagnostic workup.

CONCLUSION

Familiarization with the most common and specific patterns of muscular involvement in inherited myopathies is very important for radiologists and neurologists, as imaging plays a significant role in diagnosis and follow-up of these patients.

KEY POINTS

• Imaging is an increasingly important tool for diagnosis and follow-up in the setting of inherited myopathies. • Knowledge of the most common imaging patterns of muscle involvement in inherited myopathies is valuable for both radiologists and neurologists. • In this review, we present imaging-based algorithms that can help in the diagnostic workup of myopathies.

Overview of MR Image Segmentation Strategies in Neuromuscular Disorders

Neuromuscular disorders are rare diseases for which few therapeutic strategies currently exist. Assessment of therapeutic strategies efficiency is limited by the lack of biomarkers sensitive to the slow progression of neuromuscular diseases (NMD). Magnetic resonance imaging (MRI) has emerged as a tool of choice for the development of qualitative scores for the study of NMD. The recent emergence of quantitative MRI has enabled to provide quantitative biomarkers more sensitive to the evaluation of pathological changes in muscle tissue. However, in order to extract these biomarkers from specific regions of interest, muscle segmentation is mandatory. The time-consuming aspect of manual segmentation has limited the evaluation of these biomarkers on large cohorts. In recent years, several methods have been proposed to make the segmentation step automatic or semi-automatic. The purpose of this study was to review these methods and discuss their reliability, reproducibility, and limitations in the context of NMD. A particular attention has been paid to recent deep learning methods, as they have emerged as an effective method of image segmentation in many other clinical contexts.

The value of qualitative muscle MRI in the diagnostic procedures of myopathies: a biopsy-controlled study in 191 patients

Background and aims

The role of muscle magnetic resonance imaging (MRI) in the diagnostic procedures of myopathies is still controversially discussed. The current study was designed to analyze the status of qualitative muscle MRI, electromyography (EMG), and muscle biopsy in different cases of clinically suspected myopathy.

Methods

A total of 191 patients (male: n = 112, female: n = 79) with suspected myopathy who all received muscle MRI, EMG, and muscle biopsy for diagnostic reasons were studied, with the same location of biopsy and muscle MRI (either upper or lower extremities or paravertebral muscles). Muscle MRIs were analyzed using standard rating protocols by two different raters independently.

Results

Diagnostic findings according to biopsy results and genetic testing were as follow: non-inflammatory myopathy: n = 65, inflammatory myopathy (myositis): n = 51, neurogenic: n = 18, unspecific: n = 23, and normal: n = 34. The majority of patients showed myopathic changes in the EMG. Edema, atrophy, muscle fatty replacement, and contrast medium enhancement (CM uptake) in MRI were observed across all final diagnostic groups. Only 30% of patients from the myositis group (n = 15) showed CM uptake.

Discussion and conclusion

The study provides guidance in the definition of the impact of muscle MRI in suspected myopathy: despite being an important diagnostic tool, qualitative MRI findings could not distinguish different types of neuromuscular diagnostic groups in comparison with the gold standard histopathologic diagnosis and/or genetic testing. The results suggest that neither muscle edema nor gadolinium enhancement are able to secure a diagnosis of myositis. The current results do not support qualitative MRI as aiding in the diagnostic distinction of various myopathies. Quantitative muscle MRI is, however, useful in the diagnostic procedure of a suspected neuromuscular disease, especially with regard to assessing progression of a chronic myopathy by quantification of the degree of atrophy and fatty replacement and in exploring patterns of muscle group involvements in certain genetic myopathies.

Global versus individual muscle segmentation to assess quantitative MRI-based fat fraction changes in neuromuscular diseases

OBJECTIVES

Magnetic resonance imaging (MRI) constitutes a powerful outcome measure in neuromuscular disorders, yet there is a broad diversity of approaches in data acquisition and analysis. Since each neuromuscular disease presents a specific pattern of muscle involvement, the recommended analysis is assumed to be the muscle-by-muscle approach. We, therefore, performed a comparative analysis of different segmentation approaches, including global muscle segmentation, to determine the best strategy for evaluating disease progression.

METHODS

In 102 patients (21 immune-mediated necrotizing myopathy/IMNM, 21 inclusion body myositis/IBM, 10 GNE myopathy/GNEM, 19 Duchenne muscular dystrophy/DMD, 12 dysferlinopathy/DYSF, 7 limb-girdle muscular dystrophy/LGMD2I, 7 Pompe disease, 5 spinal muscular atrophy/SMA), two MRI scans were obtained at a 1-year interval in thighs and lower legs. Regions of interest (ROIs) were drawn in individual muscles, muscle groups, and the global muscle segment. Standardized response means (SRMs) were determined to assess sensitivity to change in fat fraction (ΔFat%) in individual muscles, muscle groups, weighted combinations of muscles and muscle groups, and in the global muscle segment.

RESULTS

Global muscle segmentation gave high SRMs for ΔFat% in thigh and lower leg for IMNM, DYSF, LGMD2I, DMD, SMA, and Pompe disease, and only in lower leg for GNEM and thigh for IBM.

CONCLUSIONS

Global muscle segment Fat% showed to be sensitive to change in most investigated neuromuscular disorders. As compared to individual muscle drawing, it is a faster and an easier approach to assess disease progression. The use of individual muscle ROIs, however, is still of interest for exploring selective muscle involvement.

KEY POINTS

• MRI-based evaluation of fatty replacement in muscles is used as an outcome measure in the assessment of 1-year disease progression in 8 different neuromuscular diseases. • Different segmentation approaches, including global muscle segmentation, were evaluated for determining 1-year fat fraction changes in lower limb skeletal muscles. • Global muscle segment fat fraction has shown to be sensitive to change in lower leg and thigh in most of the investigated neuromuscular diseases.

Skeletal muscle magnetic resonance imaging in Pompe disease

Pompe disease is characterized by a deficiency of acid alpha-glucosidase that results in muscle weakness and a variable degree of disability. There is an approved therapy based on enzymatic replacement that has modified disease progression. Several reports describing muscle magnetic resonance imaging (MRI) features of Pompe patients have been published. Most of the studies have focused on late-onset Pompe disease (LOPD) and identified a characteristic pattern of muscle involvement useful for the diagnosis. In addition, quantitative MRI studies have shown a progressive increase in fat in skeletal muscles of LOPD over time and they are increasingly considered a good tool to monitor progression of the disease. The studies performed in infantile-onset Pompe disease patients have shown less consistent changes. Other more sophisticated muscle MRI sequences, such as diffusion tensor imaging or glycogen spectroscopy, have also been used in Pompe patients and have shown promising results.

MRI in Neuromuscular Diseases: An Emerging Diagnostic Tool and Biomarker for Prognosis and Efficacy

There is an unmet need to identify biomarkers sensitive to change in rare, slowly progressive neuromuscular diseases. Quantitative magnetic resonance imaging (MRI) of muscle may offer this opportunity, as it is noninvasive and can be carried out almost independent of patient cooperation and disease severity. Muscle fat content correlates with muscle function in neuromuscular diseases, and changes in fat content precede changes in function, which suggests that muscle MRI is a strong biomarker candidate to predict prognosis and treatment efficacy. In this paper, we review the evidence suggesting that muscle MRI may be an important biomarker for diagnosis and to monitor change in disease severity. ANN NEUROL 2020;88:669-681.