Fat fraction distribution in lower limb muscles of patients with CMT1A: A quantitative MRI study

Comparison of Manual vs Artificial Intelligence-Based Muscle MRI Segmentation for Evaluating Disease Progression in Patients With CMT1A

BACKGROUND AND OBJECTIVES

Intramuscular fat fraction (FF), assessed with quantitative MRI (qMRI), has emerged as one of the few responsive outcome measures in CMT1A patients. The main limitation for its use in future therapeutic trials is the time required for the manual segmentation of individual muscles. This study aimed to evaluate the accuracy and responsiveness of a fully automatic artificial intelligence (AI)-based segmentation pipeline to assess disease progression in a cohort of CMT1A patients over 1 year.

METHODS

Twenty CMT1A patients were included in this observational, prospective, longitudinal study. FF was measured twice a year apart using qMRI in the lower limbs. Individual muscle segmentation was performed fully automatically using a trained convolutional neural network with or without human quality check (QC). The corresponding results were compared with those obtained by fully manual (FM) segmentation using the Dice similarity coefficient (DSC). FF progression and its standardized response mean (SRM) were also computed in individual muscles over the single central slice and a 3D volume to define the most sensitive region of interest.

RESULTS

AI-based segmentation showed excellent DSC values (>0.90). Significant global FF progression was observed at thigh (+0.71% ± 1.28%; p = 0.016) and leg (+1.73% ± 2.88%, p = 0.007) levels, similarly to that calculated using the FM technique (p = 0.363 and p = 0.634). FF progression of each individual muscle was comparable when computed from either the central slice or the 3D volume. The best SRM value (0.70) was obtained for the FF progression computed using the AI-based technique with human QC in the 3D volume at the leg level. The time required for fully automatic segmentation using AI with a QC was 10 hours for the entire data set compared with 90 hours for the FM.

DISCUSSION

qMRI combined with AI-based segmentation can be considered as a process ready for assessing longitudinal FF changes in CMT1A patients. Given the slow FF progression at a thigh level and the large heterogeneity between muscles and individuals, FF should be quantified from a 3D volume at the leg level for longitudinal analyses. A QC performed after the AI-based segmentation is still advised given the increased SRM value.

Clinical Characteristics of Charcot-Marie-Tooth Disease Type 4J

BACKGROUND AND OBJECTIVES

Charcot-Marie-Tooth disease type 4J (CMT4J) is caused by autosomal recessive variants in the Factor-Induced Gene 4 (FIG4) gene. Recent preclinical work has demonstrated the feasibility of adeno-associated virus serotype 9-FIG4 gene therapy. This study aimed to further characterize the CMT4J phenotype and evaluate feasibility of validated CMT-related outcome measures for future clinical trials.

METHODS

This cross-sectional study enrolled children and adults with genetically confirmed CMT4J, with 2 documented disease-causing variants in the FIG4 gene. Patients were recruited through the Inherited Neuropathy Consortium network. Disease severity was assessed using standardized CMT-specific outcome measures and exploratory biomarkers including muscle MRI fat fraction, electrophysiology, and neurofilament light chain levels. Descriptive statistics and correlation analyses were conducted to explore relationships between variables.

RESULTS

We recruited a total of 19 patients, including 14 pediatric patients (mean age 10.9 ± 3.9 years) and 5 adults (mean age 40.0 ± 13.9 years). The most frequent symptoms were gross motor delay and distal more than proximal muscle weakness, which were observed in 14 of 19 patients. The most common non-neuromuscular symptoms were cognitive and respiratory deficits, each seen in 8 of 19 patients. We denoted asymmetric weakness in 2 patients and nonuniform slowing of conduction velocities in 6 patients. Charcot-Marie-Tooth Disease Pediatric Scale (CMTPedS), Pediatric Quality of Life Inventory, and Vineland Adaptive Behavior Scale scores were affected in most patients. We observed a significant positive correlation between neurofilament light chain levels and CMTPedS, but the study was underpowered to observe a correlation between CMTPedS and MRI fat fraction.

DISCUSSION

We obtained baseline clinical and biomarker data in a broad cohort with CMT4J in pediatric and adult patients. Motor delay, muscle weakness, and respiratory and cognitive difficulties were the most common clinical manifestations of CMT4J. Many patients had nerve conduction studies with nonuniform slowing, and 2 had an asymmetric pattern of muscle weakness. We observed that the neurofilament light chain levels correlated with the CMTPedS in the pediatric population. This study showed feasibility of clinical outcomes including CMTPedS in assessment of disease severity in the pediatric patient population and provided baseline characteristics of exploratory biomarkers, neurofilament light chain levels, and muscle MRI fat fraction. The coronavirus disease 2019 pandemic affected some of the visits, resulting in a reduced number of some of the assessments.

One-Year Longitudinal Assessment of Patients With CMT1A Using Quantitative MRI

BACKGROUND AND OBJECTIVES

Intramuscular fat fraction (FF) assessed using quantitative MRI (qMRI) has emerged as one of the few responsive outcome measures in CMT1A suitable for future clinical trials. This study aimed to identify the relevance of multiple qMRI biomarkers for tracking longitudinal changes in CMT1A and to assess correlations between MRI metrics and clinical parameters.

METHODS

qMRI was performed in CMT1A patients at 2 time points, a year apart, and various metrics were extracted from 3-dimensional volumes of interest at thigh and leg levels. A semiautomated segmentation technique was used, enabling the analysis of central slices and a larger 3D muscle volume. Metrics included proton density (PD), magnetization transfer ratio (MTR), and intramuscular FF. The sciatic and tibial nerves were also assessed. Disease severity was gauged using Charcot Marie Tooth Neurologic Score (CMTNSv2), Charcot Marie Tooth Examination Score, Overall Neuropathy Limitation Scale scores, and Medical Research Council (MRC) muscle strength.

RESULTS

Twenty-four patients were included. FF significantly rose in the 3D volume at both thigh (+1.04% ± 2.19%, p = 0.041) and leg (+1.36% ± 1.87%, p = 0.045) levels. The 3D analyses unveiled a length-dependent gradient in FF, ranging from 22.61% ± 10.17% to 26.17% ± 10.79% at the leg level. There was noticeable variance in longitudinal changes between muscles: +3.17% ± 6.86% (p = 0.028) in the tibialis anterior compared with 0.37% ± 4.97% (p = 0.893) in the gastrocnemius medialis. MTR across the entire thigh volume showed a significant decline between the 2 time points -2.75 ± 6.58 (p = 0.049), whereas no significant differences were noted for the 3D muscle volume and PD. No longitudinal changes were observed in any nerve metric. Potent correlations were identified between FF and primary clinical measures: CMTNSv2 (ρ = 0.656; p = 0.001) and MRC in the lower limbs (ρ = -0.877; p < 0.001).

DISCUSSION

Our results further support that qMRI is a promising tool for following up longitudinal changes in CMT1A patients, FF being the paramount MRI metric for both thigh and leg regions. It is crucial to scrutinize the postimaging data extraction methods considering that annual changes are minimal (around +1.5%). Given the varied FF distribution, the existence of a length-dependent gradient, and the differential fatty involution across muscles, 3D volume analysis appeared more suitable than single slice analysis.

Quantified fat fraction as biomarker assessing disease severity in rare Charcot-Marie-Tooth subtypes

Objective

Charcot-Marie-Tooth (CMT) disease is the most common inherited neuromuscular disorder. Multi-echo Dixon MRI technique is a highly sensitive method for quantifying muscle fatty infiltration, which may provide excellent value for the assessment of CMT. Due to the rareness of the disease, its use in CMT disease has been rarely evaluated, especially in subtypes.

Methods

Thirty-four CMT1 patients, 25 CMT2 patients, and 10 healthy controls were recruited. All of the recruited CMT1 patients are CMT1A with PMP22 duplication. Among CMT2 patients, 7 patients are CMT2A with MFN2 mutation, and 7 patients have SORD mutations. Multi-echo Dixon MRI imaging was performed. The fat fractions (FFs) of 5 muscle compartments of the leg were measured at proximal, middle, and distal levels by two specialized musculoskeletal radiologists. Comparisons between CMT1, CMT2, and genetically defined subtypes were conducted.

Results

A proximal-distal gradient (27.6 ± 15.9, 29.9 ± 19.7, and 40.5 ± 21.4, p = 0.015) with a peroneal predominance (p = 0.001) in fat distribution was observed in CMT1. Significant differences in the soleus muscle FFs at proximal (19.1 ± 14.7 vs. 34.8 ± 25.1, p = 0.034) and medial levels (23.5 ± 21 vs. 38.0 ± 25.6, p = 0.044) were observed between CMT1 and CMT2 patients. Between PMP2 duplication and MFN2 mutation group, a significant difference in the soleus muscle FF was also observed (23.5 ± 21.0 vs. 54.7 ± 20.2, p = 0.039). Prominent correlations of calf muscle FFs with functional scores were observed.

Discussion

Multi-echo Dixon MRI imaging is a valuable tool for assessing disease severity in CMT. The difference in patterns of fatty infiltration of CMT subtypes is first reported, which could provide references when making targeted training plans.

The Impact of Fatty Infiltration on MRI Segmentation of Lower Limb Muscles in Neuromuscular Diseases: A Comparative Study of Deep Learning Approaches

BACKGROUND

Deep learning methods have been shown to be useful for segmentation of lower limb muscle MRIs of healthy subjects but, have not been sufficiently evaluated on neuromuscular disease (NDM) patients.

PURPOSE

Evaluate the influence of fat infiltration on convolutional neural network (CNN) segmentation of MRIs from NMD patients.

STUDY TYPE

Retrospective study.

SUBJECTS

Data were collected from a hospital database of 67 patients with NMDs and 14 controls (age: 53 ± 17 years, sex: 48 M, 33 F). Ten individual muscles were segmented from the thigh and six from the calf (20 slices, 200 cm section).

FIELD STRENGTH/SEQUENCE

A 1.5 T. Sequences: 2D T1 -weighted fast spin echo. Fat fraction (FF): three-point Dixon 3D GRE, magnetization transfer ratio (MTR): 3D MT-prepared GRE, T2: 2D multispin-echo sequence.

ASSESSMENT

U-Net 2D, U-Net 3D, TransUNet, and HRNet were trained to segment thigh and leg muscles (101/11 and 95/11 training/validation images, 10-fold cross-validation). Automatic and manual segmentations were compared based on geometric criteria (Dice coefficient [DSC], outlier rate, absence rate) and reliability of measured MRI quantities (FF, MTR, T2, volume).

STATISTICAL TESTS

Bland-Altman plots were chosen to describe agreement between manual vs. automatic estimated FF, MTR, T2 and volume. Comparisons were made between muscle populations with an FF greater than 20% (G20+) and lower than 20% (G20-).

RESULTS

The CNNs achieved equivalent results, yet only HRNet recognized every muscle in the database, with a DSC of 0.91 ± 0.08, and measurement biases reaching -0.32% ± 0.92% for FF, 0.19 ± 0.77 for MTR, -0.55 ± 1.95 msec for T2, and - 0.38 ± 3.67 cm3 for volume. The performances of HRNet, between G20- and G20+ decreased significantly.

DATA CONCLUSION

HRNet was the most appropriate network, as it did not omit any muscle. The accuracy obtained shows that CNNs could provide fully automated methods for studying NMDs. However, the accuracy of the methods may be degraded on the most infiltrated muscles (>20%).

EVIDENCE LEVEL

4.

TECHNICAL EFFICACY

Stage 1.

Quantification of muscle involvement in familial amyloid polyneuropathy using MRI

BACKGROUND AND PURPOSE

Transthyretin familial amyloid polyneuropathy (TTR-FAP) is a rare genetic disease with autosomal-dominant inheritance. In this study, we aimed to quantify fatty infiltration (fat fraction [FF]) and magnetization transfer ratio (MTR) in individual muscles of patients with symptomatic and asymptomatic TTR-FAP using magnetic resonance imaging. Secondarily, we aimed to assess correlations with clinical and electrophysiological variables.

METHODS

A total of 39 patients with a confirmed mutation in the TTR gene (25 symptomatic and 14 asymptomatic) and 14 healthy volunteers were included. A total of 16 muscles were manually delineated in the nondominant lower limb from T1-weighted anatomical images. The corresponding masks were propagated on the MTR and FF maps. Detailed neurological and electrophysiological examinations were conducted in each group.

RESULTS

The MTR was decreased (42.6 AU; p = 0.001) and FF was elevated (14%; p = 0.003) in the lower limbs of the symptomatic group, with preferential posterior and lateral involvement. In the asymptomatic group, elevated FF was quantified in the gastrocnemius lateralis muscle (11%; p = 0.021). FF was significantly correlated with disease duration (r = 0.49, p = 0.015), neuropathy impairment score for the lower limb (r = 0.42, p = 0.041), Overall Neuropathy Limitations Scale score (r = 0.49, p = 0.013), polyneuropathy disability score (r = 0.57, p = 0.03) and the sum of compound muscle action potential (r = 0.52, p = 0.009). MTR was strongly correlated to FF (r = 0.78, p < 0.0001), and a few muscles with an FF within the normal range had a reduced MTR.

CONCLUSION

These observations suggest that FF and MTR could be interesting biomarkers in TTR-FAP. In asymptomatic patients, FF in the gastrocnemius lateralis muscle could be a good indicator of the transition from an asymptomatic to a symptomatic form of the disease. MTR could be an early biomarker of muscle alterations.

CMT1A current gene therapy approaches and promising biomarkers

Charcot-Marie-Tooth neuropathies (CMT) constitute a group of common but highly heterogeneous, non-syndromic genetic disorders affecting predominantly the peripheral nervous system. CMT type 1A (CMT1A) is the most frequent type and accounts for almost ~50% of all diagnosed CMT cases. CMT1A results from the duplication of the peripheral myelin protein 22 (PMP22) gene. Overexpression of PMP22 protein overloads the protein folding apparatus in Schwann cells and activates the unfolded protein response. This leads to Schwann cell apoptosis, dys- and de- myelination and secondary axonal degeneration, ultimately causing neurological disabilities. During the last decades, several different gene therapies have been developed to treat CMT1A. Almost all of them remain at the pre-clinical stage using CMT1A animal models overexpressing PMP22. The therapeutic goal is to achieve gene silencing, directly or indirectly, thereby reversing the CMT1A genetic mechanism allowing the recovery of myelination and prevention of axonal loss. As promising treatments are rapidly emerging, treatment-responsive and clinically relevant biomarkers are becoming necessary. These biomarkers and sensitive clinical evaluation tools will facilitate the design and successful completion of future clinical trials for CMT1A.

Associations between lower extremity muscle fat fraction and motor performance in myotonic dystrophy type 2: A pilot study

INTRODUCTION/AIMS

Although muscle structure measures from magnetic resonance imaging (MRI) have been used to assess disease severity in muscular dystrophies, little is known about how these measures are affected in myotonic dystrophy type 2 (DM2). We aim to characterize lower extremity muscle fat fraction (MFF) as a potential biomarker of disease severity, and evaluate its relationship with motor performance in DM2.

METHODS

3-Tesla MRIs were obtained from nine patients with DM2 and six controls using a T1W-Dixon protocol. To calculate MFF, muscle volumes were segmented from proximal, middle, and distal regions of the thigh and calf. Associations between MFF and motor performance were calculated using Spearman's correlations (ρ).

RESULTS

Mean age of DM2 participants was 62 ± 11 y (89% female), and mean symptom duration was 20 ± 12 y. Compared to controls, the DM2 group had significantly higher MFF in the thigh and the calf segments (p-value = .002). The highest MFF at the thigh in DM2 was located in the posterior compartment (39.7 ± 12.9%) and at the calf was the lateral compartment (31.5 ± 8.7%). In the DM2 group, we found a strong correlation between the posterior thigh MFF and the 6-min walk test (ρ = -.90, p-value = .001). The lateral calf MFF was also strongly correlated with the step test (ρ = -0.82, p-value = .006).

DISCUSSION

Our pilot data suggest a potential correlation between lower extremity MFF and some motor performance tests in DM2. Longitudinal studies with larger sample sizes are required to validate MFF as a marker of disease severity in DM2.

Muscle quantitative MRI as a novel biomarker in hereditary transthyretin amyloidosis with polyneuropathy: a cross-sectional study

BACKGROUND

The development of reproducible and sensitive outcome measures has been challenging in hereditary transthyretin (ATTRv) amyloidosis. Recently, quantification of intramuscular fat by magnetic resonance imaging (MRI) has proven as a sensitive marker in patients with other genetic neuropathies. The aim of this study was to investigate the role of muscle quantitative MRI (qMRI) as an outcome measure in ATTRv.

METHODS

Calf- and thigh-centered multi-echo T2-weighted spin-echo and gradient-echo sequences were obtained in patients with ATTRv amyloidosis with polyneuropathy (n = 24) and healthy controls (n = 12). Water T2 (wT2) and fat fraction (FF) were calculated. Neurological assessment was performed in all ATTRv subjects. Quantitative MRI parameters were correlated with clinical and neurophysiological measures of disease severity.

RESULTS

Quantitative imaging revealed significantly higher FF in lower limb muscles in patients with ATTRv amyloidosis compared to controls. In addition, wT2 was significantly higher in ATTRv patients. There was prominent involvement of the posterior compartment of the thighs. Noticeably, FF and wT2 did not exhibit a length-dependent pattern in ATTRv patients. MRI biomarkers correlated with previously validated clinical outcome measures, Polyneuropathy Disability scoring system, Neuropathy Impairment Score (NIS) and NIS-lower limb, and neurophysiological parameters of axonal damage regardless of age, sex, treatment and TTR mutation.

CONCLUSIONS

Muscle qMRI revealed significant difference between ATTRv and healthy controls. MRI biomarkers showed high correlation with clinical and neurophysiological measures of disease severity making qMRI as a promising tool to be further investigated in longitudinal studies to assess its role at monitoring onset, progression, and therapy efficacy for future clinical trials on this treatable condition.

Hereditary neuropathy

The hereditary neuropathies, collectively referred as Charcot-Marie-Tooth disease (CMT) and related disorders, are heterogeneous genetic peripheral nerve disorders that collectively comprise the commonest inherited neurological disease with an estimated prevalence of 1:2500 individuals. The field of hereditary neuropathies has made significant progress in recent years with respect to both gene discovery and treatment as a result of next-generation sequencing (NGS) approach. These investigations which have identified over 100 causative genes and new mutations have made the classification of CMT even more challenging. Despite so many different mutated genes, the majority of CMT forms share a similar clinical phenotype, and due to this phenotypic homogeneity, genetic testing in CMT is increasingly being performed through the use of NGS panels. The majority of patients still have a mutation in one the four most common genes (PMP22 duplication-CMT1A, MPZ-CMT1B, GJB1-CMTX1, and MFN2-CMT2A). This chapter focuses primarily on these four forms and their potential therapeutic approaches.

Magnetic resonance imaging-based lower limb muscle evaluation in Charcot-Marie-Tooth disease type 1A patients and its correlation with clinical data

We aimed to derive comprehensive MRI parameters that reflect intramuscular fat infiltration severity for designated lower extremity levels, based on semiquantitative analyses in Charcot-Marie-Tooth disease type 1A (CMT1A) patients. We reviewed lower extremity MRIs of 116 CMT1A patients. Intramuscular fat infiltration grading using the Mercuri scale was performed for the non-dominant lower extremity at three levels (proximal, mid, and distal) for the thigh and at two levels (proximal and distal) for the lower leg. Based on MRI results, the following parameters were calculated for each level and for entire muscles: fat infiltration proportion (FIP), significant fat infiltration proportion (SigFIP), and severe fat infiltration proportion (SevFIP). The relationships between the MRI parameters and clinical data were evaluated using Spearman’s correlation analysis. FIP, SigFIP, and SevFIP measured for entire muscles significantly correlated with Charcot-Marie-Tooth Neuropathy Score (p < 0.001), functional disability scale (p < 0.001), 10-m walk test time (p = 0.0003, 0.0010, and 0.0011), and disease duration (p < 0.001). Similar correlations were demonstrated for FIP, SigFIP, and SevFIP acquired from the lower leg. Our MRI parameters obtained through semiquantitative analyses of muscles significantly correlated with clinical parameters in CMT1A patients, suggesting their potential applicability as imaging markers for clinical severity.

Skeletal Muscle Involvement Pattern of Hereditary Transthyretin Amyloidosis: A Study Based on Muscle MRI

Objects

This study was intended to explore the characteristics of muscle magnetic resonance imaging (MRI) of patients with hereditary transthyretin amyloidosis (ATTRv amyloidosis) prospectively.

Methods

The clinical data of 20 patients with ATTRv amyloidosis at our hospital between July 2020 and August 2021 were analyzed. MRI of lower limbs including calf muscles was performed in all these 20 patients and MRI of thigh muscles was performed in 16 of them.

Results

The mean age of the 20 patients with ATTRv amyloidosis was 44.2 years (ranging from 26 to 60) whose mean duration of weakness was 23.3 ± 23.0 (ranging from 0 to 84) months. All the patients presented with polyneuropathy, and 18 of them with weakness in their lower limbs. Muscle involvement was selective in these patients with ATTRv amyloidosis. The posterior group of muscles was heavily fatty, and the soleus muscle was the most heavily involved. The proportion of fatty infiltration scores at the calf level was higher than at the thigh level with paired comparison for most patients. Three of these patients had more severely fatty infiltration of muscles at the thigh level. The fatty infiltration of posterior compartments at the calf level was highly consistent with neuropathy impairment scores of lower limbs (weakness), the strength of ankle plantar flexion muscles, and the amplitude of the compound muscle action potential of the tibial nerve.

Conclusions

It was found that the pattern of muscle fatty infiltration was consistent with a distal-to-proximal gradient on the whole and that proximal involvements in MRI of lower limbs in some patients could also be observed. Selective fatty infiltration of muscles of posterior compartments and fatty infiltration of the soleus muscle might be typical of ATTRv amyloidosis.

New Keys to Early Diagnosis: Muscle Echogenicity, Nerve Ultrasound Patterns, Electrodiagnostic, and Clinical Parameters in 150 Patients with Hereditary Polyneuropathies

Hereditary neuropathies are of variable genotype and phenotype. With upcoming therapies, there is urgent need for early disease recognition and outcome measures. High-resolution nerve and muscle ultrasound is a dynamic, non-invasive, well-established tool in the field of inflammatory and traumatic neuropathies. In this study, we defined nerve and muscle ultrasound parameters as recognition and progression markers in 150 patients with genetically confirmed hereditary neuropathies, including Charcot-Marie-Tooth (CMT) disease (CMT1A, n = 55; other CMT1/4, n = 28; axonal CMT, n = 15; CMTX, n = 15), hereditary neuropathy with liability to pressure palsies (HNPP, n = 16), hereditary transthyretin-amyloidosis (ATTRv, n = 14), and Fabry's disease (n = 7). The CMT1A, followed by the CMT1/4 group, had the most homogeneous enlargement of the nerve cross-sectional areas (CSA) in the ultrasound pattern sum (UPSS) and homogeneity score. Entrapment scores were highest in HNPP, ATTRv amyloidosis, and Fabry's disease patients. In demyelinating neuropathies, the CSA correlated inversely with nerve conduction studies. The muscle echo intensity was significantly highest in the clinically most affected muscles, which was independent from the underlying disease cause and correlated with muscle strength and disease duration. Further correlations were seen with combined clinical (CMTES-2) and electrophysiological (CMTNS-2) scores of disease severity. We conclude that nerve ultrasound is a helpful tool to distinguish different types of hereditary neuropathies by pattern recognition, whereas muscle ultrasound is an objective parameter for disease severity. The implementation of neuromuscular ultrasound might enrich diagnostic procedures both in clinical routines and research.

3D Automated Segmentation of Lower Leg Muscles Using Machine Learning on a Heterogeneous Dataset

Quantitative MRI combines non-invasive imaging techniques to reveal alterations in muscle pathophysiology. Creating muscle-specific labels manually is time consuming and requires an experienced examiner. Semi-automatic and fully automatic methods reduce segmentation time significantly. Current machine learning solutions are commonly trained on data from healthy subjects using homogeneous databases with the same image contrast. While yielding high Dice scores (DS), those solutions are not applicable to different image contrasts and acquisitions. Therefore, the aim of our study was to evaluate the feasibility of automatic segmentation of a heterogeneous database. To create a heterogeneous dataset, we pooled lower leg muscle images from different studies with different contrasts and fields-of-view, containing healthy controls and diagnosed patients with various neuromuscular diseases. A second homogenous database with uniform contrasts was created as a subset of the first database. We trained three 3D-convolutional neuronal networks (CNN) on those databases to test performance as compared to manual segmentation. All networks, training on heterogeneous data, were able to predict seven muscles with a minimum average DS of 0.75. U-Net performed best when trained on the heterogeneous dataset (DS: 0.80 ± 0.10, AHD: 0.39 ± 0.35). ResNet and DenseNet yielded higher DS, when trained on a heterogeneous dataset (both DS: 0.86), as compared to a homogeneous dataset (ResNet DS: 0.83, DenseNet DS: 0.76). In conclusion, a CNN trained on a heterogeneous dataset achieves more accurate labels for predicting a heterogeneous database of lower leg muscles than a CNN trained on a homogenous dataset. We propose that a large heterogeneous database is needed, to make automated segmentation feasible for different kinds of image acquisitions.

Quantitative assessment of muscle echogenicity in Charcot-Marie-Tooth disease type 1A by automatic thresholding methods

OBJECTIVE

To investigate the utility of automatic thresholding methods for quantitative muscle echogenicity assessment as a marker of disease severity in Charcot-Marie-Tooth disease type 1A (CMT1A).

METHODS

Muscle ultrasound was performed in 15 CMT1A patients and 7 healthy controls. Muscle echogenicity of six limb muscles in each subject was assessed by 16 automatic thresholding methods and conventional grey-scale analysis. Echogenicity of each method in CMT1A patients was compared with that in controls. A correlation between the echogenicity and CMT neuropathy score (CMTNS) was also analysed in CMT1A patients.

RESULTS

Significant differences in mean echogenicity of the 6 muscles between CMT1A patients and controls were found both in grey-scale analysis (p < 0.01) and 11 of the 16 automatic thresholding methods (p < 0.05 in each method). In CMT1A patients, mean echogenicity of the 6 muscles was positively correlated with CMTNS in 8 of the 16 automatic thresholding methods, but not in grey-scale analysis.

CONCLUSION

Automatic thresholding methods can be used to detect the difference in muscle echogenicity between CMT1A patients and controls. Echogenicity parameters correlate with the disease severity.

SIGNIFICANCE

Quantitative muscle echogenicity assessment by automatic thresholding methods shows potential as a surrogate marker of disease progression in CMT1A.

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.

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.

Semi-Automatic MRI Muscle Volumetry to Diagnose and Monitor Hereditary and Acquired Polyneuropathies

With emerging treatment approaches, it is crucial to correctly diagnose and monitor hereditary and acquired polyneuropathies. This study aimed to assess the validity and accuracy of magnet resonance imaging (MRI)-based muscle volumetry.Using semi-automatic segmentations of upper- and lower leg muscles based on whole-body MRI and axial T1-weighted turbo spin-echo sequences, we compared and correlated muscle volumes, and clinical and neurophysiological parameters in demyelinating Charcot-Marie-Tooth disease (CMT) (n = 13), chronic inflammatory demyelinating polyneuropathy (CIDP) (n = 27), and other neuropathy (n = 17) patients.The muscle volumes of lower legs correlated with foot dorsiflexion strength (p < 0.0001), CMT Neuropathy Score 2 (p < 0.0001), early gait disorders (p = 0.0486), and in CIDP patients with tibial nerve conduction velocities (p = 0.0092). Lower (p = 0.0218) and upper (p = 0.0342) leg muscles were significantly larger in CIDP compared to CMT patients. At one-year follow-up (n = 15), leg muscle volumes showed no significant decrease.MRI muscle volumetry is a promising method to differentiate and characterize neuropathies in clinical practice.

Texture analysis using T1-weighted images for muscles in Charcot-Marie-Tooth disease patients and volunteers

OBJECTIVES

To explore whether texture features using T1-weighted images correlate with fat fraction, and whether they differ between Charcot-Marie-Tooth (CMT) disease patients and volunteers.

METHODS

The institutional review board approved this retrospective study, and the requirement for informed consent was waived; data of eighteen CMT patients and eighteen healthy volunteers from a previous study was used. Texture features of the muscles including mean, standard deviation (SD), skewness, kurtosis, and entropy of the signal intensity were derived from T1-weighted images. Spearman's correlation analysis was used to assess the relationship between texture features and fat fraction measured by 3D multiple gradient echo Dixon-based sequence. Mann-Whitney U test was used to compare the texture features between CMT patients and volunteers. Intraobserver and interobserver agreements for the texture features were assessed using the intraclass correlation coefficient.

RESULTS

The SD (ρ = 0.256, p < 0.001) and entropy (ρ = 0.263, p < 0.001) were significantly and positively correlated with fat fraction; skewness (ρ = - 0.110, p = 0.027) and kurtosis (ρ = - 0.149, p = 0.003) were significantly and inversely correlated with fat fraction. The CMT patients showed a significantly higher SD (63.45 vs. 49.26; p < 0.001), skewness (1.06 vs. 0.56; p < 0.001), kurtosis (4.00 vs. 1.81; p < 0.001), and entropy (3.20 vs. 3.02; p < 0.001) than did the volunteers. Intraobserver and interobserver agreements were almost perfect for mean, SD, and entropy.

CONCLUSIONS

Texture features using T1-weighted images correlated with fat fraction and differed between CMT patients and volunteers.

KEY POINTS

• Standard deviation and entropy of muscles derived from T1-weighted images were significantly and positively correlated with the muscle fat fraction. • Mean, standard deviation, and entropy were considered highly reliable in muscle analyses. • Texture features may have the potential to diagnose early stage of intramuscular fatty infiltration.

Quantitative assessment of sciatic nerve changes in Charcot-Marie-Tooth type 1A patients using magnetic resonance neurography

BACKGROUND AND PURPOSE

Nerve tissue alterations have rarely been quantified in Charcot-Marie-Tooth type 1A (CMT1A) patients. The aim of the present study was to quantitatively assess the magnetic resonance imaging (MRI) anomalies of the sciatic and tibial nerves in CMT1A disease using quantitative neurography MRI. It was also intended to seek for correlations with clinical variables.

METHODS

Quantitative neurography MRI was used in order to assess differences in nerve volume, proton density and magnetization transfer ratio in the lower limbs of CMT1A patients and healthy controls. Disease severity was evaluated using the Charcot-Marie-Tooth Neuropathy Score version 2, Charcot-Marie-Tooth examination scores and Overall Neuropathy Limitations Scale scores. Electrophysiological measurements were performed in order to assess the compound motor action potential and the Motor Unit Number Index. Clinical impairment was evaluated using muscle strength measurements and Charcot-Marie-Tooth examination scores.

RESULTS

A total of 32 CMT1A patients were enrolled and compared to 13 healthy subjects. The 3D nerve volume, magnetization transfer ratio and proton density were significantly different in CMT1A patients for the whole sciatic and tibial nerve volume. The sciatic nerve volume was significantly correlated with the whole set of clinical scores whereas no correlation was found between the tibial nerve volume and the clinical scores.

CONCLUSION

Nerve injury could be quantified in vivo using quantitative neurography MRI and the corresponding biomarkers were correlated with clinical disability in CMT1A patients. The sensitivity of the selected metrics will have to be assessed through repeated measurements over time during longitudinal studies to evaluate structural nerve changes under treatment.