Volume measurements of individual muscles in human quadriceps femoris using atlas-based segmentation approaches

Characteristics of paraspinal muscle degeneration in patients with adult degenerative scoliosis

INTRODUCTION

Adult degenerative scoliosis (ADS) is a 3D deformity that greatly affects the quality of life of patients and is closely related to the quality of paraspinal muscles (PSMs), but the specific degenerative characteristics have not been described.

METHODS

This study included ADS patients who were first diagnosed in our hospital from 2018 to 2022. Muscle volume (MV) and fat infiltration (FI) of PSM were measured by 3D reconstruction, and spinal parameters were assessed by X-ray. The values of convex side (CV) and concave side (CC) were compared.

RESULTS

Fifty patients were enrolled with a mean age of 64.1 ± 5.8 years old. There were significant differences in MV, FI, and Cobb angle between male and female groups. The MV of MF and PS on the CC was significantly larger than that on the CV. In the apex and the segments above the apex, the FI of the MF on the CC is greater than the CV, and in the CV of the segment below the apex, the FI of the MF is greater than the CC. Besides, there was a significant positive correlation between the FI and Cobb angle in the MF of the CC-CV.

CONCLUSION

There were significant differences in the MV and FI of PSM on both sides of the spine in ADS patients. It was determined that the PSM of ADS showed different degrees of degeneration in different levels of the lumbar spine and were positively correlated with Cobb angle.

Validity and Reliability of 3-D Ultrasound Imaging to Measure Hamstring Muscle and Tendon Volumes

OBJECTIVE

The validity and reliability of 3-D ultrasound (US) in estimation of muscle and tendon volume was assessed in a very limited number of muscles that can be easily immersed. The objective of the present study was to assess the validity and reliability of muscle volume measurements for all hamstring muscle heads and gracilis (GR), as well as tendon volume for the semitendinosus (ST) and GR using freehand 3-D US.

METHODS

Three-dimensional US acquisitions were performed for 13 participants in two distinct sessions on separate days, in addition to one session dedicated to magnetic resonance imaging (MRI). Volumes of ST, semimembranosus (SM), biceps femoris short (BFsh) and long (BFlh) heads, and GR muscles and from the tendon from semitendinosus (STtd) and gracilis (GRtd) were collected.

RESULTS

The bias and the 95% confidence intervals of 3-D US compared with MRI ranged from -1.9 mL (-0.8%) to 1.2 mL (1.0%) for muscle volume and from 0.01 mL (0.2%) to -0.03 mL (-2.6%) for tendon volume. For muscle volume assessed using 3-D US, intraclass correlation coefficients (ICCs) ranged from 0.98 (GR) to 1.00, and coefficients of variation (CV) from 1.1% (SM) to 3.4% (BFsh). For tendon volume, ICCs were 0.99, and CVs between 3.2% (STtd) and 3.4% (GRtd).

CONCLUSION

Three-dimensional US can provide a valid and reliable inter-day measurement of hamstrings and GR for both muscle and tendon volumes. In the future, this technique could be used as an outcome for strengthening interventions and potentially in clinical environments.

Deep learning-based automatic pipeline for quantitative assessment of thigh muscle morphology and fatty infiltration

PURPOSE

Fast and accurate thigh muscle segmentation from MRI is important for quantitative assessment of thigh muscle morphology and composition. A novel deep learning (DL) based thigh muscle and surrounding tissues segmentation model was developed for fully automatic and reproducible cross-sectional area (CSA) and fat fraction (FF) quantification and tested in patients at 10 years after anterior cruciate ligament reconstructions.

METHODS

A DL model combining UNet and DenseNet was trained and tested using manually segmented thighs from 16 patients (32 legs). Segmentation accuracy was evaluated using Dice similarity coefficients (DSC) and average symmetric surface distance (ASSD). A UNet model was trained for comparison. These segmentations were used to obtain CSA and FF quantification. Reproducibility of CSA and FF quantification was tested with scan and rescan of six healthy subjects.

RESULTS

The proposed UNet and DenseNet had high agreement with manual segmentation (DSC >0.97, ASSD < 0.24) and improved performance compared with UNet. For hamstrings of the operated knee, the automated pipeline had largest absolute difference of 6.01% for CSA and 0.47% for FF as compared to manual segmentation. In reproducibility analysis, the average difference (absolute) in CSA quantification between scan and rescan was better for the automatic method as compared with manual segmentation (2.27% vs. 3.34%), whereas the average difference (absolute) in FF quantification were similar.

CONCLUSIONS

The proposed method exhibits excellent accuracy and reproducibility in CSA and FF quantification compared with manual segmentation and can be used in large-scale patient studies.

Automated MRI quantification of volumetric per-muscle fat fractions in the proximal leg of patients with muscular dystrophies

Muscular dystrophies (MD) are a class of rare genetic diseases resulting in progressive muscle weakness affecting specific muscle groups, depending on the type of disease. Disease progression is characterized by the gradual replacement of muscle tissue by fat, which can be assessed with fat-sensitive magnetic resonance imaging (MRI) and objectively evaluated by quantifying the fat fraction percentage (FF%) per muscle. Volumetric quantification of fat replacement over the full 3D extent of each muscle is more precise and potentially more sensitive than 2D quantification in few selected slices only, but it requires an accurate 3D segmentation of each muscle individually, which is time consuming when this has to be performed manually for a large number of muscles. A reliable, largely automated approach for 3D muscle segmentation is thus needed to facilitate the adoption of fat fraction quantification as a measure of MD disease progression in clinical routine practice, but this is challenging due to the variable appearance of the images and the ambiguity in the discrimination of the contours of adjacent muscles, especially when the normal image contrast is affected and diminished by the fat replacement. To deal with these challenges, we used deep learning to train AI-models to segment the muscles in the proximal leg from knee to hip in Dixon MRI images of healthy subjects as well as patients with MD. We demonstrate state-of-the-art segmentation results of all 18 muscles individually in terms of overlap (Dice score, DSC) with the manual ground truth delineation for images of cases with low fat infiltration (mean overall FF%: 11.3%; mean DSC: 95.3% per image, 84.4-97.3% per muscle) as well as with medium and high fat infiltration (mean overall FF%: 44.3%; mean DSC: 89.0% per image, 70.8-94.5% per muscle). In addition, we demonstrate that the segmentation performance is largely invariant to the field of view of the MRI scan, is generalizable to patients with different types of MD and that the manual delineation effort to create the training set can be drastically reduced without significant loss of segmentation quality by delineating only a subset of the slices.

Magnetic resonance imaging assessments for knee segmentation and their use in combination with machine/deep learning as predictors of early osteoarthritis diagnosis and prognosis

Knee osteoarthritis (OA) is a prevalent and disabling disease that can develop over decades. This disease is heterogeneous and involves structural changes in the whole joint, encompassing multiple tissue types. Detecting OA before the onset of irreversible changes is crucial for early management, and this could be achieved by allowing knee tissue visualization and quantifying their changes over time. Although some imaging modalities are available for knee structure assessment, magnetic resonance imaging (MRI) is preferred. This narrative review looks at existing literature, first on MRI-developed approaches for evaluating knee articular tissues, and second on prediction using machine/deep-learning-based methodologies and MRI as input or outcome for early OA diagnosis and prognosis. A substantial number of MRI methodologies have been developed to assess several knee tissues in a semi-quantitative and quantitative fashion using manual, semi-automated and fully automated systems. This dynamic field has grown substantially since the advent of machine/deep learning. Another active area is predictive modelling using machine/deep-learning methodologies enabling robust early OA diagnosis/prognosis. Moreover, incorporating MRI markers as input/outcome in such predictive models is important for a more accurate OA structural diagnosis/prognosis. The main limitation of their usage is the ability to move them in rheumatology practice. In conclusion, MRI knee tissue determination and quantification provide early indicators for individuals at high risk of developing this disease or for patient prognosis. Such assessment of knee tissues, combined with the development of models/tools from machine/deep learning using, in addition to other parameters, MRI markers for early diagnosis/prognosis, will maximize opportunities for individualized risk assessment for use in clinical practice permitting precision medicine. Future efforts should be made to integrate such prediction models into open access, allowing early disease management to prevent or delay the OA outcome.

Deformable image registration based on single or multi-atlas methods for automatic muscle segmentation and the generation of augmented imaging datasets

Muscle segmentation is a process relied upon to gather medical image-based muscle characterisation, useful in directly assessing muscle volume and geometry, that can be used as inputs to musculoskeletal modelling pipelines. Manual or semi-automatic techniques are typically employed to segment the muscles and quantify their properties, but they require significant manual labour and incur operator repeatability issues. In this study an automatic process is presented, aiming to segment all lower limb muscles from Magnetic Resonance (MR) imaging data simultaneously using three-dimensional (3D) deformable image registration (single inputs or multi-atlas). Twenty-three of the major lower limb skeletal muscles were segmented from five subjects, with an average Dice similarity coefficient of 0.72, and average absolute relative volume error (RVE) of 12.7% (average relative volume error of -2.2%) considering the optimal subject combinations. The multi-atlas approach showed slightly better accuracy (average DSC: 0.73; average RVE: 1.67%). Segmented MR imaging datasets of the lower limb are not widely available in the literature, limiting the potential of new, probabilistic methods such as deep learning to be used in the context of muscle segmentation. In this work, Non-linear deformable image registration is used to generate 69 manually checked, segmented, 3D, artificial datasets, allowing access for future studies to use these new methods, with a large amount of reliable reference data.

Intra-operator Repeatability of Manual Segmentations of the Hip Muscles on Clinical Magnetic Resonance Images

The manual segmentation of muscles on magnetic resonance images is the gold standard procedure to reconstruct muscle volumes from medical imaging data and extract critical information for clinical and research purposes. (Semi)automatic methods have been proposed to expedite the otherwise lengthy process. These, however, rely on manual segmentations. Nonetheless, the repeatability of manual muscle volume segmentations performed on clinical MRI data has not been thoroughly assessed. When conducted, volumetric assessments often disregard the hip muscles. Therefore, one trained operator performed repeated manual segmentations (n = 3) of the iliopsoas (n = 34) and gluteus medius (n = 40) muscles on coronal T1-weighted MRI scans, acquired on 1.5 T scanners on a clinical population of patients elected for hip replacement surgery. Reconstructed muscle volumes were divided in sub-volumes and compared in terms of volume variance (normalized variance of volumes - nVV), shape (Jaccard Index-JI) and surface similarity (maximal Hausdorff distance-HD), to quantify intra-operator repeatability. One-way repeated measures ANOVA (or equivalent) tests with Bonferroni corrections for multiple comparisons were conducted to assess statistical significance. For both muscles, repeated manual segmentations were highly similar to one another (nVV: 2-6%, JI > 0.78, HD < 15 mm). However, shape and surface similarity were significantly lower when muscle extremities were included in the segmentations (e.g., iliopsoas: HD -12.06 to 14.42 mm, P < 0.05). Our findings show that the manual segmentation of hip muscle volumes on clinical MRI scans provides repeatable results over time. Nonetheless, extreme care should be taken in the segmentation of muscle extremities.

Ultrasound and magnetic resonance imaging as diagnostic tools for sarcopenia in immune-mediated rheumatic diseases (IMRDs)

Sarcopenia is characterized by loss of muscle mass, altered muscle composition, fat and fibrous tissue infiltration, and abnormal innervation, especially in older individuals with immune-mediated rheumatic diseases (IMRDs). Several techniques for measuring muscle mass, strength, and performance have emerged in recent decades. The portable dynamometer and gait speed represent the most frequently used tools for the evaluation of muscle strength and physical efficiency, respectively. Aside from dual-energy, X-ray, absorptiometry, and bioelectrical impedance analysis, ultrasound (US) and magnetic resonance imaging (MRI) techniques appear to have a potential role in evaluating muscle mass and composition. US and MRI have been shown to accurately identify sarcopenic biomarkers such as inflammation (edema), fatty infiltration (myosteatosis), alterations in muscle fibers, and muscular atrophy in patients with IMRDs. US is a low-cost, easy-to-use, and safe imaging method for assessing muscle mass, quality, architecture, and biomechanical function. This review summarizes the evidence for using US and MRI to assess sarcopenia.

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

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

CMC-Net: 3D calf muscle compartment segmentation with sparse annotation

Accurate 3D segmentation of calf muscle compartments in volumetric MR images is essential to diagnose as well as assess progression of muscular diseases. Recently, good segmentation performance was achieved using state-of-the-art deep learning approaches, which, however, require large amounts of annotated data for training. Considering that obtaining sufficiently large medical image annotation datasets is often difficult, time-consuming, and requires expert knowledge, minimizing the necessary sizes of expert-annotated training datasets is of great importance. This paper reports CMC-Net, a new deep learning framework for calf muscle compartment segmentation in 3D MR images that selects an effective small subset of 2D slices from the 3D images to be labelled, while also utilizing unannotated slices to facilitate proper generalization of the subsequent training steps. Our model consists of three parts: (1) an unsupervised method to select the most representative 2D slices on which expert annotation is performed; (2) ensemble model training employing these annotated as well as additional unannotated 2D slices; (3) a model-tuning method using pseudo-labels generated by the ensemble model that results in a trained deep network capable of accurate 3D segmentations. Experiments on segmentation of calf muscle compartments in 3D MR images show that our new approach achieves good performance with very small annotation ratios, and when utilizing full annotation, it outperforms state-of-the-art full annotation segmentation methods. Additional experiments on a 3D MR thigh dataset further verify the ability of our method in segmenting leg muscle groups with sparse annotation.

Reference values for volume, fat content and shape of the hip abductor muscles in healthy individuals from Dixon MRI

Healthy hip abductor muscles are a good indicator of a healthy hip and an active lifestyle, as they are greatly involved in human daily activities. Fatty infiltration and muscle atrophy are associated with loss of strength, loss of mobility and hip disease. However, these variables have not been widely studied in this muscle group. We aimed to characterize the hip abductor muscles in a group of healthy individuals to establish reference values for volume, intramuscular fat content and shape of this muscle group. To achieve this, we executed a cross-sectional study using Dixon MRI scans of 51 healthy subjects. We used an automated segmentation method to label GMAX, GMED, GMIN and TFL muscles, measured normalized volume (NV) using lean body mass, fat fraction (FF) and lean muscle volume for each subject and computed non-parametric statistics for each variable grouped by sex and age. We measured these variables for each axial slice and created cross-sectional area and FF axial profiles for each muscle. Finally, we generated sex-specific atlases with FF statistical images. We measured median (IQR) NV values of 12.6 (10.8-13.8), 6.3 (5.6-6.7), 1.6 (1.4-1.7) and 0.8 (0.6-1.0) cm³ /kg for GMAX, GMED, GMIN and TFL, and median (IQR) FF values of 12.3 (10.1-15.9)%, 9.8 (8.6-11.2)%, 10.0 (9.0-12.0)% and 10.2 (7.8-13.5)% respectively. FF values were significantly higher for females for the four muscles (p < 0.01), but there were no significant differences between the two age groups. When comparing individual muscles, we observed a significantly higher FF in GMAX than in the other muscles. The reported novel reference values and axial profiles for volume and FF of the hip abductors, together with male and female atlases, are tools that could potentially help to quantify and detect early the deteriorating effects of hip disease or sarcopenia.

Intramuscular fat in gluteus maximus for different levels of physical activity

We aimed to determine if gluteus maximus (GMAX) fat infiltration is associated with different levels of physical activity. Identifying and quantifying differences in the intramuscular fat content of GMAX in subjects with different levels of physical activity can provide a new tool to evaluate hip muscles health. This was a cross-sectional study involving seventy subjects that underwent Dixon MRI of the pelvis. The individuals were divided into four groups by levels of physical activity, from low to high: inactive patients due to hip pain; and low, medium and high physical activity groups of healthy subjects (HS) based on hours of exercise per week. We estimated the GMAX intramuscular fat content for each subject using automated measurements of fat fraction (FF) from Dixon images. The GMAX volume and lean volume were also measured and normalized by lean body mass. The effects of body mass index (BMI) and age were included in the statistical analysis. The patient group had a significantly higher FF than the three groups of HS (median values of 26.2%, 17.8%, 16.7% and 13.7% respectively, p < 0.001). The normalized lean volume was significantly larger in the high activity group compared to all the other groups (p < 0.001, p = 0.002 and p = 0.02). Employing a hierarchical linear regression analysis, we found that hip pain, low physical activity, female gender and high BMI were statistically significant predictors of increased GMAX fat infiltration.

Deep learning for automatic segmentation of thigh and leg muscles

Objective

In this study we address the automatic segmentation of selected muscles of the thigh and leg through a supervised deep learning approach.

Material and methods

The application of quantitative imaging in neuromuscular diseases requires the availability of regions of interest (ROI) drawn on muscles to extract quantitative parameters. Up to now, manual drawing of ROIs has been considered the gold standard in clinical studies, with no clear and universally accepted standardized procedure for segmentation. Several automatic methods, based mainly on machine learning and deep learning algorithms, have recently been proposed to discriminate between skeletal muscle, bone, subcutaneous and intermuscular adipose tissue. We develop a supervised deep learning approach based on a unified framework for ROI segmentation.

Results

The proposed network generates segmentation maps with high accuracy, consisting in Dice Scores ranging from 0.89 to 0.95, with respect to “ground truth” manually segmented labelled images, also showing high average performance in both mild and severe cases of disease involvement (i.e. entity of fatty replacement).

Discussion

The presented results are promising and potentially translatable to different skeletal muscle groups and other MRI sequences with different contrast and resolution.

Deep generative models for automated muscle segmentation in computed tomography scanning

Accurate gluteus medius (GMd) volume evaluation may aid in the analysis of muscular atrophy states and help gain an improved understanding of patient recovery via rehabilitation. However, the segmentation of muscle regions in GMd images for cubic muscle volume assessment is time-consuming and labor-intensive. This study automated GMd-region segmentation from the computed tomography (CT) images of patients diagnosed with hip osteoarthritis using deep learning and evaluated the segmentation accuracy. To this end, 5250 augmented pairs of training data were obtained from five participants, and a conditional generative adversarial network was used to identify the relationships between the image pairs. Using the preserved test datasets, the results of automatic segmentation with the trained deep learning model were compared to those of manual segmentation in terms of the dice similarity coefficient (DSC), volume similarity (VS), and shape similarity (MS). As observed, the average DSC values for automatic and manual segmentations were 0.748 and 0.812, respectively, with a significant difference (p < 0.0001); the average VS values were 0.247 and 0.203, respectively, with no significant difference (p = 0.069); and the average MS values were 1.394 and 1.156, respectively, with no significant difference (p = 0.308). The GMd volumes obtained by automatic and manual segmentation were 246.2 cm³ and 282.9 cm³, respectively. The noninferiority of the DSC obtained by automatic segmentation was verified against that obtained by manual segmentation. Accordingly, the proposed GAN-based automatic GMd-segmentation technique is confirmed to be noninferior to manual segmentation. Therefore, the findings of this research confirm that the proposed method not only reduces time and effort but also facilitates accurate assessment of the cubic muscle volume.

Regional variation of thigh muscle fat infiltration in patients with neuromuscular diseases compared to healthy controls

Background

Chemical shift encoding-based water-fat magnetic resonance imaging (CSE-MRI) measures a quantitative biomarker: the proton density fat fraction (PDFF). The aim was to assess regional and proximo-distal PDFF variations at the thigh in patients with myotonic dystrophy type 2 (DM2), limb-girdle muscular dystrophy type 2A (LGMD2A), and late-onset Pompe disease (LOPD) as compared to healthy controls.

Methods

Seven patients (n=2 DM2, n=2 LGMD2A, n=3 LOPD) and 20 controls were recruited. A 3D-spoiled gradient echo sequence was used to scan the thigh musculature. Muscles were manually segmented to generate mean muscle PDFF.

Results

In all three disease entities, there was an increase in muscle fat replacement compared to healthy controls. However, within each disease group, there were patients with a shorter time since symptom onset that only showed mild PDFF elevation (range, 10% to 20%) compared to controls (P≤0.05), whereas patients with a longer period since symptom onset showed a more severe grade of fat replacement with a range of 50% to 70% (P<0.01). Increased PDFF of around 5% was observed for vastus medialis, semimembranosus and gracilis muscles in advanced compared to early DM2. LGMD2A_1 showed an early disease stage with predominantly mild PDFF elevations over all muscles and levels (10.9%±7.1%) compared to controls. The quadriceps, gracilis and biceps femoris muscles showed the highest difference between LGMD2A_1 with 5 years since symptom onset (average PDFF 11.1%±6.9%) compared to LGMD2A_2 with 32 years since symptom onset (average PDFF 66.3%±6.3%). For LOPD patients, overall PDFF elevations were observed in all major hip flexors and extensors (range, 25.8% to 30.8%) compared to controls (range, 1.7% to 2.3%, P<0.05). Proximal-to-distal PDFF highly varied within and between diseases and within controls. The intra-reader reliability was high (reproducibility coefficient ≤2.19%).

Conclusions

By quantitatively measuring muscle fat infiltration at the thigh, we identified candidate muscles for disease monitoring due to their gradual PDFF elevation with longer disease duration. Regional variation between proximal, central, and distal muscle PDFF was high and is important to consider when performing longitudinal MRI follow-ups in the clinical setting or in longitudinal studies.

Evaluation of Amyotrophic Lateral Sclerosis-Induced Muscle Degeneration Using Magnetic Resonance-Based Relaxivity Contrast Imaging (RCI)

(1) Background: This work characterizes the sensitivity of magnetic resonance-based Relaxivity Contrast Imaging (RCI) to Amyotrophic Lateral Sclerosis (ALS)-induced changes in myofiber microstructure. Transverse Relaxivity at Tracer Equilibrium (TRATE), an RCI-based parameter, was evaluated in the lower extremities of ALS patients and healthy subjects. (2) Methods: In this IRB-approved study, 23 subjects (12 ALS patients and 11 healthy controls) were scanned at 3T (Philips, The Netherlands). RCI data were obtained during injection of a gadolinium-based contrast agent. TRATE, fat fraction and T₂ measures, were compared in five muscle groups of the calf muscle, between ALS and control populations. TRATE was also evaluated longitudinally (baseline and 6 months) and was compared to clinical measures, namely ALS Functional Rating Scale (ALSFRS-R) and Hand-Held Dynamometry (HHD), in a subset of the ALS population. (3) Results: TRATE was significantly lower (p < 0.001) in ALS-affected muscle than in healthy muscle in all muscle groups. Fat fraction differences between ALS and healthy muscle were statistically significant for the tibialis anterior (p = 0.01), tibialis posterior (p = 0.004), and peroneus longus (p = 0.02) muscle groups but were not statistically significant for the medial (p = 0.07) and lateral gastrocnemius (p = 0.06) muscles. T₂ differences between ALS and healthy muscle were statistically significant for the tibialis anterior (p = 0.004), peroneus longus (p = 0.004) and lateral gastrocnemius (p = 0.03) muscle groups but were not statistically significant for the tibialis posterior (p = 0.06) and medial gastrocnemius (p = 0.07) muscles. Longitudinally, TRATE, averaged over all patients, decreased by 28 ± 16% in the tibialis anterior, 47 ± 18% in the peroneus longus, 25 ± 19% in the tibialis posterior, 29 ± 14% in the medial gastrocnemius and 35 ± 18% in the lateral gastrocnemius muscles between two timepoints. ALSFRS-R scores were stable in two of four ALS patients. HHD scores decreased in three of four ALS patients. (4) Conclusion: RCI-based TRATE was shown to consistently differentiate ALS-affected muscle from healthy muscle and also provide a quantitative measure of longitudinal muscle degeneration.

Semiautomatic quantification of abdominal wall muscles deformations based on dynamic MRI image registration

Quantitative analysis of abdominal organs motion and deformation is crucial to better understand biomechanical alterations undermining respiratory, digestive or perineal pathophysiology. In particular, biomechanical characterization of the antero-lateral abdominal wall is central in the diagnosis of abdominal muscle deficiency. Here, we present a dedicated semiautomatic dynamic MRI postprocessing method enabling the quantification of spatial and temporal deformations of the antero-lateral abdominal wall muscles. Ten healthy participants were imaged during a controlled breathing session at the L3-L4 disc level using real-time dynamic MRI at 3 T. A coarse feature-tracking step allowed the selection of the inhalation cycle of maximum abdominal excursion. Over this image series, the described method combines (1) a supervised 2D+t segmentation procedure of the abdominal wall muscles, (2) the quantification of muscle deformations based on masks registration, and (3) the mapping of deformations within muscle subzones leveraging a dedicated automatic parcellation. The supervised 2D+t segmentation (1) provided an accurate segmentation of the abdominal wall muscles throughout maximum inhalation with a 0.95 ± 0.03 Dice similarity coefficient (DSC) value and a 2.3 ± 0.7 mm Hausdorff distance value while requiring only manual segmentation of 20% of the data. The robustness of the deformation quantification (2) was indicated by high indices of correspondence between the registered source mask and the target mask (0.98 ± 0.01 DSC value and 2.1 ± 1.5 mm Hausdorff distance value). Parcellation (3) enabled the distinction of muscle substructures that are anatomically relevant but could not be distinguished based on image contrast. The present genuine postprocessing method provides a quantitative analytical frame that could be used in further studies for a better understanding of abdominal wall deformations in physiological and pathological situations.

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.

Use of artificial intelligence in the imaging of sarcopenia: A narrative review of current status and perspectives

Sarcopenia is a muscle disease which previously was associated only with aging, but in recent days it has been gaining more attention for its predictive value in a vast range of conditions and its potential link with overall health. Up to this point, evaluating sarcopenia with imaging methods has been time-consuming and dependent on the skills of the physician. The solution for this problem may be found in artificial intelligence, which may assist radiologists in repetitive tasks such as muscle segmentation and body-composition analysis. The major aim of this review was to find and present the current status and future perspectives of artificial intelligence in the imaging of sarcopenia. We searched the PubMed database to find articles concerning the use of artificial intelligence in diagnostic imaging and especially in body-composition analysis in the context of sarcopenia. We found that artificial-intelligence systems could potentially help with evaluating sarcopenia and better predicting outcomes in a vast range of clinical situations, which could get us closer to the true era of precision medicine.

Fully automated 3D segmentation of MR-imaged calf muscle compartments: Neighborhood relationship enhanced fully convolutional network

Automated segmentation of individual calf muscle compartments from 3D magnetic resonance (MR) images is essential for developing quantitative biomarkers for muscular disease progression and its prediction. Achieving clinically acceptable results is a challenging task due to large variations in muscle shape and MR appearance. In this paper, we present a novel fully convolutional network (FCN) that utilizes contextual information in a large neighborhood and embeds edge-aware constraints for individual calf muscle compartment segmentations. An encoder-decoder architecture is used to systematically enlarge convolution receptive field and preserve information at all resolutions. Edge positions derived from the FCN output muscle probability maps are explicitly regularized using kernel-based edge detection in an end-to-end optimization framework. Our method was evaluated on 40 T1-weighted MR images of 10 healthy and 30 diseased subjects by fourfold cross-validation. Mean DICE coefficients of 88.00-91.29% and mean absolute surface positioning errors of 1.04-1.66 mm were achieved for the five 3D muscle compartments.

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.

The optimisation of deep neural networks for segmenting multiple knee joint tissues from MRIs

Automated semantic segmentation of multiple knee joint tissues is desirable to allow faster and more reliable analysis of large datasets and to enable further downstream processing e.g. automated diagnosis. In this work, we evaluate the use of conditional Generative Adversarial Networks (cGANs) as a robust and potentially improved method for semantic segmentation compared to other extensively used convolutional neural network, such as the U-Net. As cGANs have not yet been widely explored for semantic medical image segmentation, we analysed the effect of training with different objective functions and discriminator receptive field sizes on the segmentation performance of the cGAN. Additionally, we evaluated the possibility of using transfer learning to improve the segmentation accuracy. The networks were trained on i) the SKI10 dataset which comes from the MICCAI grand challenge "Segmentation of Knee Images 2010″, ii) the OAI ZIB dataset containing femoral and tibial bone and cartilage segmentations of the Osteoarthritis Initiative cohort and iii) a small locally acquired dataset (Advanced MRI of Osteoarthritis (AMROA) study) consisting of 3D fat-saturated spoiled gradient recalled-echo knee MRIs with manual segmentations of the femoral, tibial and patellar bone and cartilage, as well as the cruciate ligaments and selected peri-articular muscles. The Sørensen-Dice Similarity Coefficient (DSC), volumetric overlap error (VOE) and average surface distance (ASD) were calculated for segmentation performance evaluation. DSC ≥ 0.95 were achieved for all segmented bone structures, DSC ≥ 0.83 for cartilage and muscle tissues and DSC of ≈0.66 were achieved for cruciate ligament segmentations with both cGAN and U-Net on the in-house AMROA dataset. Reducing the receptive field size of the cGAN discriminator network improved the networks segmentation performance and resulted in segmentation accuracies equivalent to those of the U-Net. Pretraining not only increased segmentation accuracy of a few knee joint tissues of the fine-tuned dataset, but also increased the network's capacity to preserve segmentation capabilities for the pretrained dataset. cGAN machine learning can generate automated semantic maps of multiple tissues within the knee joint which could increase the accuracy and efficiency for evaluating joint health.

The Role of Imaging Biomarkers in the Assessment of Sarcopenia

BACKGROUND

The diagnosis of sarcopenia through clinical assessment has some limitations. The literature advises studies that include objective markers along with clinical assessment in order to improve the sensitivity and specificity of current diagnostic criteria. The decrease of muscle quality precedes the loss of quantity, so we studied the role magnetic resonance imaging biomarkers as indicators of the quantity and quality of muscle in sarcopenia patients.

METHODS

a cross-sectional analysis was performed to analyze what MR-derived imaging parameters correlate better with sarcopenia diagnostic criteria in women of 70 years of age and over (independent walking and community-dwelling women who were sarcopenic in accordance with EWGSOP criteria with muscle mass adjusted to Spanish population were chosen).

RESULTS

The study included 26 women; 81 ± 8 years old. A strong correlation was obtained between cineanthropometric variables (BMI; thigh perimeter and fat mass) and imaging biomarkers (muscle/fat ratio, fatty infiltration, muscle T2*, water diffusion coefficient, and proton density fat fraction) with coefficients around 0.7 (absolute value).

CONCLUSIONS

Knowing the correlation of clinical parameters and imaging-derived muscle quality indicators can help to identify older women at risk of developing sarcopenia at an early stage. This may allow taking preventive actions to decrease disability, morbidity, and mortality in sarcopenia patients.

Automated measurement of fat infiltration in the hip abductors from Dixon magnetic resonance imaging

PURPOSE

Intramuscular fat infiltration is a dynamic process, in response to exercise and muscle health, which can be quantified by estimating fat fraction (FF) from Dixon MRI. Healthy hip abductor muscles are a good indicator of a healthy hip and an active lifestyle as they have a fundamental role in walking. The automated measurement of the abductors' FF requires the challenging task of segmenting them. We aimed to design, develop and evaluate a multi-atlas based method for automated measurement of fat fraction in the main hip abductor muscles: gluteus maximus (GMAX), gluteus medius (GMED), gluteus minimus (GMIN) and tensor fasciae latae (TFL).

METHOD

We collected and manually segmented Dixon MR images of 10 healthy individuals and 7 patients who underwent MRI for hip problems. Twelve of them were selected to build an atlas library used to implement the automated multi-atlas segmentation method. We compared the FF in the hip abductor muscles for the automated and manual segmentations for both healthy and patients groups. Measures of average and spread were reported for FF for both methods. We used the root mean square error (RMSE) to quantify the method accuracy. A linear regression model was used to explain the relationship between FF for automated and manual segmentations.

RESULTS

The automated median (IQR) FF was 20.0(16.0-26.4) %, 14.3(10.9-16.5) %, 15.5(13.9-18.6) % and 16.2(13.5-25.6) % for GMAX, GMED, GMIN and TFL respectively, with a FF RMSE of 1.6%, 0.8%, 2.1%, 2.7%. A strong linear correlation (R² = 0.93, p < .001, m = 0.99) was found between the FF from automated and manual segmentations. The mean FF was higher in patients than in healthy subjects.

CONCLUSION

The automated measurement of FF of hip abductor muscles from Dixon MRI had good agreement with FF measurements from manually segmented images. The method was accurate for both healthy and patients groups.

Healthy versus pathological learning transferability in shoulder muscle MRI segmentation using deep convolutional encoder-decoders

Fully-automated segmentation of pathological shoulder muscles in patients with musculo-skeletal diseases is a challenging task due to the huge variability in muscle shape, size, location, texture and injury. A reliable automatic segmentation method from magnetic resonance images could greatly help clinicians to diagnose pathologies, plan therapeutic interventions and predict interventional outcomes while eliminating time consuming manual segmentation. The purpose of this work is three-fold. First, we investigate the feasibility of automatic pathological shoulder muscle segmentation using deep learning techniques, given a very limited amount of available annotated pediatric data. Second, we address the learning transferability from healthy to pathological data by comparing different learning schemes in terms of model generalizability. Third, extended versions of deep convolutional encoder-decoder architectures using encoders pre-trained on non-medical data are proposed to improve the segmentation accuracy. Methodological aspects are evaluated in a leave-one-out fashion on a dataset of 24 shoulder examinations from patients with unilateral obstetrical brachial plexus palsy and focus on 4 rotator cuff muscles (deltoid, infraspinatus, supraspinatus and subscapularis). The most accurate segmentation model is partially pre-trained on the large-scale ImageNet dataset and jointly exploits inter-patient healthy and pathological annotated data. Its performance reaches Dice scores of 82.4%, 82.0%, 71.0% and 82.8% for deltoid, infraspinatus, supraspinatus and subscapularis muscles. Absolute surface estimation errors are all below 83 mm² except for supraspinatus with 134.6 mm². The contributions of our work offer new avenues for inferring force from muscle volume in the context of musculo-skeletal disorder management.

A Knowledge-Based Modality-Independent Technique for Concurrent Thigh Muscle Segmentation: Applicable to CT and MR Images

The mass of the lower extremity muscles is a clinically significant metric. Manual segmentation of these muscles is a time-consuming task. Most of the segmentation methods for the thigh muscles are based on statistical models and atlases which need manually segmented datasets. The goal of this work is an automatic segmentation of the thigh muscles with only one initial segmented slice. A new automatic method is proposed for concurrent individual thigh muscles segmentation using a hybrid level set method and anatomical information of the muscles. In the proposed method, the muscle regions are extracted by the Fast and Robust Fuzzy C-Means Clustering (FRFCM) method, and then a contour is determined for each muscle which changes according to the muscle shape variation through its length. The anatomical information is used to control the contours variations and to refine the final boundaries. The method was validated by 22 CT datasets. The average dice similarity coefficient (DSC) of the method for individual muscle segmentation with one and two initial slices were 89.29 ± 2.59 (%) and 91.77 ± 1.87 (%), respectively. Also, the average symmetric surface distances (ASSDs) were 0.93 ± 0.29 mm and 0.64 ± 0.18 mm. Furthermore, applying to ten MRI datasets, the average DSC and ASSD for muscles were 90.9 ± 2.61 (%) and 0.71 ± 0.33 mm, respectively. The quantitative and intuitive results of the proposed method show the effectiveness of this method in segmentation of large and small muscles in CT and MR images. The consumed computation time is lower than the previous works, and this method does not need any training datasets.

Multi-atlas-based auto-segmentation for prostatic urethra using novel prediction of deformable image registration accuracy

PURPOSE

Accurate identification of the prostatic urethra and bladder can help determine dosing and evaluate urinary toxicity during intensity-modulated radiation therapy (IMRT) planning in patients with localized prostate cancer. However, it is challenging to locate the prostatic urethra in planning computed tomography (pCT). In the present study, we developed a multiatlas-based auto-segmentation method for prostatic urethra identification using deformable image registration accuracy prediction with machine learning (ML) and assessed its feasibility.

METHODS

We examined 120 patients with prostate cancer treated with IMRT. All patients underwent temporary urinary catheter placement for identification and contouring of the prostatic urethra in pCT images (ground truth). Our method comprises the following three steps: (a) select four atlas datasets from the atlas datasets using the deformable image registration (DIR) accuracy prediction model, (b) deform them by structure-based DIR, (3) and propagate urethra contour using displacement vector field calculated by the DIR. In (a), for identifying suitable datasets, we used the trained support vector machine regression (SVR) model and five feature descriptors (e.g., prostate volume) to increase DIR accuracy. This method was trained/validated using 100 patients and performance was evaluated within an independent test set of 20 patients. Fivefold cross-validation was used to optimize the hype parameters of the DIR accuracy prediction model. We assessed the accuracy of our method by comparing it with those of two others: Acostas method-based patient selection (previous study method, by Acosta et al.), and the Waterman's method (defines the prostatic urethra based on the center of the prostate, by Waterman et al.). We used the centerlines distance (CLD) between the ground truth and the predicted prostatic urethra as the evaluation index.

RESULTS

The CLD in the entire prostatic urethra was 2.09 ± 0.89 mm (our proposed method), 2.77 ± 0.99 mm (Acosta et al., P = 0.022), and 3.47 ± 1.19 mm (Waterman et al., P < 0.001); our proposed method showed the highest accuracy. In segmented CLD, CLD in the top 1/3 segment was highly improved from that of Waterman et.al. and was slightly improved from that of Acosta et.al., with results of 2.49 ± 1.78 mm (our proposed method), 2.95 ± 1.75 mm (Acosta et al., P = 0.42), and 5.76 ± 3.09 mm (Waterman et al., P < 0.001).

CONCLUSIONS

We developed a DIR accuracy prediction model-based multiatlas-based auto-segmentation method for prostatic urethra identification. Our method identified prostatic urethra with mean error of 2.09 mm, likely due to combined effects of SVR model employment in patient selection, modified atlas dataset characteristics and DIR algorithm. Our method has potential utility in prostate cancer IMRT and can replace use of temporary indwelling urinary catheters.

Automated multi-atlas segmentation of gluteus maximus from Dixon and T1-weighted magnetic resonance images

OBJECTIVE

To design, develop and evaluate an automated multi-atlas method for segmentation and volume quantification of gluteus maximus from Dixon and T1-weighted images.

MATERIALS AND METHODS

The multi-atlas segmentation method uses an atlas library constructed from 15 Dixon MRI scans of healthy subjects. A non-rigid registration between each atlas and the target, followed by majority voting label fusion, is used in the segmentation. We propose a region of interest (ROI) to standardize the measurement of muscle bulk. The method was evaluated using the dice similarity coefficient (DSC) and the relative volume difference (RVD) as metrics, for Dixon and T1-weighted target images.

RESULTS

The mean(± SD) DSC was 0.94 ± 0.01 for Dixon images, while 0.93 ± 0.02 for T1-weighted. The RVD between the automated and manual segmentation had a mean(± SD) value of 1.5 ± 4.3% for Dixon and 1.5 ± 4.8% for T1-weighted images. In the muscle bulk ROI, the DSC was 0.95 ± 0.01 and the RVD was 0.6 ± 3.8%.

CONCLUSION

The method allows an accurate fully automated segmentation of gluteus maximus for Dixon and T1-weighted images and provides a relatively accurate volume measurement in shorter times (~ 20 min) than the current gold-standard manual segmentations (2 h). Visual inspection of the segmentation would be required when higher accuracy is needed.

Exploration of New Contrasts, Targets, and MR Imaging and Spectroscopy Techniques for Neuromuscular Disease - A Workshop Report of Working Group 3 of the Biomedicine and Molecular Biosciences COST Action BM1304 MYO-MRI

Neuromuscular diseases are characterized by progressive muscle degeneration and muscle weakness resulting in functional disabilities. While each of these diseases is individually rare, they are common as a group, and a large majority lacks effective treatment with fully market approved drugs. Magnetic resonance imaging and spectroscopy techniques (MRI and MRS) are showing increasing promise as an outcome measure in clinical trials for these diseases. In 2013, the European Union funded the COST (co-operation in science and technology) action BM1304 called MYO-MRI (www.myo-mri.eu), with the overall aim to advance novel MRI and MRS techniques for both diagnosis and quantitative monitoring of neuromuscular diseases through sharing of expertise and data, joint development of protocols, opportunities for young researchers and creation of an online atlas of muscle MRI and MRS. In this report, the topics that were discussed in the framework of working group 3, which had the objective to: Explore new contrasts, new targets and new imaging techniques for NMD are described. The report is written by the scientists who attended the meetings and presented their data. An overview is given on the different contrasts that MRI can generate and their application, clinical needs and desired readouts, and emerging methods.

Automated postoperative muscle assessment of hip arthroplasty patients using multimodal imaging joint segmentation

BACKGROUND AND OBJECTIVE

In patients treated with hip arthroplasty, the muscular condition and presence of inflammatory reactions are assessed using magnetic resonance imaging (MRI). As MRI lacks contrast for bony structures, computed tomography (CT) is preferred for clinical evaluation of bone tissue and orthopaedic surgical planning. Combining the complementary information of MRI and CT could improve current clinical practice for diagnosis, monitoring and treatment planning. In particular, the different contrast of these modalities could help better quantify the presence of fatty infiltration to characterise muscular condition and assess implant failure. In this work, we combine CT and MRI for joint bone and muscle segmentation and we propose a novel Intramuscular Fat Fraction estimation method for the quantification of muscle atrophy.

METHODS

Our multimodal framework is able to segment healthy and pathological musculoskeletal structures as well as implants, and develops into three steps. First, input images are pre-processed to improve the low quality of clinically acquired images and to reduce the noise associated with metal artefact. Subsequently, CT and MRI are non-linearly aligned using a novel approach which imposes rigidity constraints on bony structures to ensure realistic deformation. Finally, taking advantage of a multimodal atlas we created for this task, a multi-atlas based segmentation delineates pelvic bones, abductor muscles and implants on both modalities jointly. From the obtained segmentation, a multimodal estimation of the Intramuscular Fat Fraction can be automatically derived.

RESULTS

Evaluation of the segmentation in a leave-one-out cross-validation study on 22 hip sides resulted in an average Dice score of 0.90 for skeletal and 0.84 for muscular structures. Our multimodal Intramuscular Fat Fraction was benchmarked on 27 different cases against a standard radiological score, showing stronger association than a single modality approach in a one-way ANOVA F-test analysis.

CONCLUSIONS

The proposed framework represents a promising tool to support image analysis in hip arthroplasty, being robust to the presence of implants and associated image artefacts. By allowing for the automated extraction of a muscle atrophy imaging biomarker, it could quantitatively inform the decision-making process about patient's management.

Clinical evaluation of fully automated thigh muscle and adipose tissue segmentation using a U-Net deep learning architecture in context of osteoarthritic knee pain

OBJECTIVE

Segmentation of thigh muscle and adipose tissue is important for the understanding of musculoskeletal diseases such as osteoarthritis. Therefore, the purpose of this work is (a) to evaluate whether a fully automated approach provides accurate segmentation of muscles and adipose tissue cross-sectional areas (CSA) compared with manual segmentation and (b) to evaluate the validity of this method based on a previous clinical study.

MATERIALS AND METHODS

The segmentation method is based on U-Net architecture trained on 250 manually segmented thighs from the Osteoarthritis Initiative (OAI). The clinical evaluation is performed on a hold-out test set bilateral thighs of 48 subjects with unilateral knee pain.

RESULTS

The segmentation time of the method is < 1 s and demonstrated high agreement with the manual method (dice similarity coeffcient: 0.96 ± 0.01). In the clinical study, the automated method shows that similar to manual segmentation (- 5.7 ± 7.9%, p < 0.001, effect size: 0.69), painful knees display significantly lower quadriceps CSAs than contralateral painless knees (- 5.6 ± 7.6%, p < 0.001, effect size: 0.73).

DISCUSSION

Automated segmentation of thigh muscle and adipose tissues has high agreement with manual segmentations and can replicate the effect size seen in a clinical study on osteoarthritic pain.

Novel stochastic framework for automatic segmentation of human thigh MRI volumes and its applications in spinal cord injured individuals

Severe spinal cord injury (SCI) leads to skeletal muscle atrophy and adipose tissue infiltration in the skeletal muscle, which can result in compromised muscle mechanical output and lead to health-related complications. In this study, we developed a novel automatic 3-D approach for volumetric segmentation and quantitative assessment of thigh Magnetic Resonance Imaging (MRI) volumes in individuals with chronic SCI as well as non-disabled individuals. In this framework, subcutaneous adipose tissue, inter-muscular adipose tissue and total muscle tissue are segmented using linear combination of discrete Gaussians algorithm. Also, three thigh muscle groups were segmented utilizing the proposed 3-D Joint Markov Gibbs Random Field model that integrates first order appearance model, spatial information, and shape model to localize the muscle groups. The accuracy of the automatic segmentation method was tested both on SCI (N = 16) and on non-disabled (N = 14) individuals, showing an overall 0.93±0.06 accuracy for adipose tissue and muscle compartments segmentation based on Dice Similarity Coefficient. The proposed framework for muscle compartment segmentation showed an overall higher accuracy compared to ANTs and STAPLE, two previously validated atlas-based segmentation methods. Also, the framework proposed in this study showed similar Dice accuracy and better Hausdorff distance measure to that obtained using DeepMedic Convolutional Neural Network structure, a well-known deep learning network for 3-D medical image segmentation. The automatic segmentation method proposed in this study can provide fast and accurate quantification of adipose and muscle tissues, which have important health and functional implications in the SCI population.

Quantifying skeletal muscle volume and shape in humans using MRI: A systematic review of validity and reliability

AIMS

The aim of this study was to report the metrological qualities of techniques currently used to quantify skeletal muscle volume and 3D shape in healthy and pathological muscles.

METHODS

A systematic review was conducted (Prospero CRD42018082708). PubMed, Web of Science, Cochrane and Scopus databases were searched using relevant keywords and inclusion/exclusion criteria. The quality of the articles was evaluated using a customized scale.

RESULTS

Thirty articles were included, 6 of which included pathological muscles. Most evaluated lower limb muscles. Partially or completely automatic and manual techniques were assessed in 10 and 24 articles, respectively. Manual slice-by-slice segmentation reliability was good-to-excellent (n = 8 articles) and validity against dissection was moderate to good(n = 1). Manual slice-by-slice segmentation was used as a gold-standard method in the other articles. Reduction of the number of manually segmented slices (n = 6) provided good to excellent validity if a sufficient number of appropriate slices was chosen. Segmentation on one slice (n = 11) increased volume errors. The Deformation of a Parametric Specific Object (DPSO) method (n = 5) decreased the number of manually-segmented slices required for any chosen level of error. Other automatic techniques combined with different statistical shape or atlas/images-based methods (n = 4) had good validity. Some particularities were highlighted for specific muscles. Except for manual slice by slice segmentation, reliability has rarely been reported.

CONCLUSIONS

The results of this systematic review help the choice of appropriate segmentation techniques, according to the purpose of the measurement. In healthy populations, techniques that greatly simplified the process of manual segmentation yielded greater errors in volume and shape estimations. Reduction of the number of manually segmented slices was possible with appropriately chosen segmented slices or with DPSO. Other automatic techniques showed promise, but data were insufficient for their validation. More data on the metrological quality of techniques used in the cases of muscle pathology are required.

Fully automatic segmentation of paraspinal muscles from 3D torso CT images via multi-scale iterative random forest classifications

PURPOSE

To develop and validate a fully automatic method for segmentation of paraspinal muscles from 3D torso CT images.

METHODS

We propose a novel learning-based method to address this challenging problem. Multi-scale iterative random forest classifications with multi-source information are employed in this study to speed up the segmentation and to improve the accuracy. Here, multi-source images include the original torso CT images and later also the iteratively estimated and refined probability maps of the paraspinal muscles. We validated our method on 20 torso CT data with associated manual segmentation. We randomly partitioned the 20 CT data into two evenly distributed groups and took one group as the training data and the other group as the test data.

RESULTS

The proposed method achieved a mean Dice coefficient of 93.0%. It took on average 46.5 s to segment a 3D torso CT image with the size ranging from [Formula: see text] voxels to [Formula: see text] voxels.

CONCLUSIONS

Our fully automatic, learning-based method can accurately segment paraspinal muscles from 3D torso CT images. It generates segmentation results that are better than those achieved by the state-of-the-art methods.

MRI adipose tissue and muscle composition analysis-a review of automation techniques

MRI is becoming more frequently used in studies involving measurements of adipose tissue and volume and composition of skeletal muscles. The large amount of data generated by MRI calls for automated analysis methods. This review article presents a summary of automated and semi-automated techniques published between 2013 and 2017. Technical aspects and clinical applications for MRI-based adipose tissue and muscle composition analysis are discussed based on recently published studies. The conclusion is that very few clinical studies have used highly automated analysis methods, despite the rapidly increasing use of MRI for body composition analysis. Possible reasons for this are that the availability of highly automated methods has been limited for non-imaging experts, and also that there is a limited number of studies investigating the reproducibility of automated methods for MRI-based body composition analysis.

Individual muscle segmentation in MR images: A 3D propagation through 2D non-linear registration approaches

Manual and automated segmentation of individual muscles in magnetic resonance images have been recognized as challenging given the high variability of shapes between muscles and subjects and the discontinuity or lack of visible boundaries between muscles. In the present study, we proposed an original algorithm allowing a semi-automatic transversal propagation of manually-drawn masks. Our strategy was based on several ascending and descending non-linear registration approaches which is similar to the estimation of a Lagrangian trajectory applied to manual masks. Using several manually-segmented slices, we have evaluated our algorithm on the four muscles of the quadriceps femoris group. We mainly showed that our 3D propagated segmentation was very accurate with an averaged Dice similarity coefficient value higher than 0.91 for the minimal manual input of only two manually-segmented slices.

Thigh muscle segmentation of chemical shift encoding-based water-fat magnetic resonance images: The reference database MyoSegmenTUM

Magnetic resonance imaging (MRI) can non-invasively assess muscle anatomy, exercise effects and pathologies with different underlying causes such as neuromuscular diseases (NMD). Quantitative MRI including fat fraction mapping using chemical shift encoding-based water-fat MRI has emerged for reliable determination of muscle volume and fat composition. The data analysis of water-fat images requires segmentation of the different muscles which has been mainly performed manually in the past and is a very time consuming process, currently limiting the clinical applicability. An automatization of the segmentation process would lead to a more time-efficient analysis. In the present work, the manually segmented thigh magnetic resonance imaging database MyoSegmenTUM is presented. It hosts water-fat MR images of both thighs of 15 healthy subjects and 4 patients with NMD with a voxel size of 3.2x2x4 mm³ with the corresponding segmentation masks for four functional muscle groups: quadriceps femoris, sartorius, gracilis, hamstrings. The database is freely accessible online at https://osf.io/svwa7/?view_only=c2c980c17b3a40fca35d088a3cdd83e2. The database is mainly meant as ground truth which can be used as training and test dataset for automatic muscle segmentation algorithms. The segmentation allows extraction of muscle cross sectional area (CSA) and volume. Proton density fat fraction (PDFF) of the defined muscle groups from the corresponding images and quadriceps muscle strength measurements/neurological muscle strength rating can be used for benchmarking purposes.

Automated muscle segmentation from CT images of the hip and thigh using a hierarchical multi-atlas method

PURPOSE

Patient-specific quantitative assessments of muscle mass and biomechanical musculoskeletal simulations require segmentation of the muscles from medical images. The objective of this work is to automate muscle segmentation from CT data of the hip and thigh.

METHOD

We propose a hierarchical multi-atlas method in which each hierarchy includes spatial normalization using simpler pre-segmented structures in order to reduce the inter-patient variability of more complex target structures.

RESULTS

The proposed hierarchical method was evaluated with 19 muscles from 20 CT images of the hip and thigh using the manual segmentation by expert orthopedic surgeons as ground truth. The average symmetric surface distance was significantly reduced in the proposed method (1.53 mm) in comparison with the conventional method (2.65 mm).

CONCLUSION

We demonstrated that the proposed hierarchical multi-atlas method improved the accuracy of muscle segmentation from CT images, in which large inter-patient variability and insufficient contrast were involved.

Muscle Atrophy Measurement as Assessment Method for Low Back Pain Patients

Low back pain is one of the most common pain disorders defined as pain, muscle tension, or stiffness localized below the costal margin and above the inferior gluteal folds, sometimes with accompanying leg pain. The meaning of the symptomatic atrophy of paraspinal muscles and some pelvic muscles has been proved. Nowadays, a need for new diagnostic tools for specific examination of low back pain patients is posited, and it has been proposed that magnetic resonance imaging assessment toward muscle atrophy may provide some additional information enabling the subclassification of that group of patients.

Muscle Near-Infrared Spectroscopy Signals versus Venous Blood Hemoglobin Oxygen Saturation in Skeletal Muscle

PURPOSE

The study aimed to examine the relationship between near-infrared spectroscopy (NIRS) signals and venous hemoglobin oxygen saturation (O2Hb%) and venous oxygen concentration (CvO2).

METHODS

Gastrocnemius muscles (GS) in six dogs were surgically isolated and pump perfused. NIRS signals were recorded, and venous blood samples were collected at constant flow rates (control flow, high flow, and low flow) at rest as well as during electrically stimulated tetanic muscle contractions at rates of one contraction per 2 s (1/2 s) and two contractions per 3 s (2/3 s). Similar data were also collected at three different inspired O2 percentages (12%, 21%, and 100%) with constant blood flow.

RESULTS

Complete data from five animals were analyzed; all data from one animal were deleted because of erratic oxy-NIRS signals. Venous O2Hb% ranged from 7.6% to 97.5% across the various experimental conditions. After the NIRS signals were normalized to the physiological range, a high linear correlation was seen between the deoxygenated heme signal (HHbMb%) and the venous O2Hb% (R = 0.92 ± 0.05), between the oxygenated heme signal (HbMbO2%) and the venous O2Hb% (R = 0.92 ± 0.03), between the HHbMb% and the CvO2 (R = 0.89 ± 0.06), and between the HbMbO2% and the CvO2 (R = 0.90 ± 0.05). The overall relationships between HHbMb%, HbMbO2%, and venous O2Hb% as well as between HHbMb%, HbMbO2%, and CvO2 were also linear and highly correlated with R values ranging from 0.81 to 0.90.

CONCLUSION

In this controlled canine muscle model, NIRS signals are highly correlated with venous O2Hb% and CvO2 across a wide range of physiological conditions. The practical application of our results is that for an individual muscle or perhaps muscle group, normalized NIRS HHbMb and HbMbO2 signals accurately reflect the mean venous O2 saturation of the interrogated muscle tissue.

Long-term follow-up of MRI changes in thigh muscles of patients with Facioscapulohumeral dystrophy: A quantitative study

Facioscapulohumeral muscular dystrophy (FSHD) is one of the most common hereditary muscular disorders. Currently FSHD has no known effective treatment and detailed data on the natural history are lacking. Determination of the efficacy of a given therapeutic approach might be difficult in FSHD given the slow and highly variable disease progression. Magnetic resonance imaging (MRI) has been widely used to qualitatively and quantitatively evaluate in vivo the muscle alterations in various neuromuscular disorders. The main aim of the present study was to investigate longitudinally the time-dependent changes occurring in thigh muscles of FSHD patients using quantitative MRI and to assess the potential relationships with the clinical findings. Thirty-five FSHD1 patients (17 females) were enrolled. Clinical assessment tools including manual muscle testing using medical research council score (MRC), and motor function measure (MFM) were recorded each year for a period ranging from 1 to 2 years. For the MRI measurements, we used a new quantitative index, i.e., the mean pixel intensity (MPI) calculated from the pixel-intensity distribution in T1 weighted images. The corresponding MPI scores were calculated for each thigh, for each compartment and for both thighs totally (MPI_total). The total mean pixel intensity (MPI_total) refers to the sum of each pixel signal intensity divided by the corresponding number of pixels. An increased MPI_total indicates both a raised fat infiltration together with a reduced muscle volume thereby illustrating disease progression. Clinical scores did not change significantly over time whereas MPI_total increased significantly from an initial averaged value of 39.6 to 41.1 with a corresponding rate of 0.62/year. While clinical scores and MPI_total measured at the start of the study were significantly related, no correlation was found between the rate of MPI_total and MRC sum score changes, MFM_total and MFM subscores. The relative rate of MPI_total change was 2.3% (0.5–4.3)/year and was significantly higher than the corresponding rates measured for MRCS 0% (0–1.7) /year and MFM_total 0% (0–2.0) /year (p = 0.000). On the basis of these results, we suggested that muscle MRI and more particularly the MPI_total index could be used as a reliable biomarker and outcome measure of disease progression. In slowly progressive myopathies such as FSHD, the MPI_total index might reveal subclinical changes, which could not be evidenced using clinical scales over a short period of time.

Validation of an active shape model-based semi-automated segmentation algorithm for the analysis of thigh muscle and adipose tissue cross-sectional areas

Objective

To validate a semi-automated method for thigh muscle and adipose tissue cross-sectional area (CSA) segmentation from MRI.

Materials and methods

An active shape model (ASM) was trained using 113 MRI CSAs from the Osteoarthritis Initiative (OAI) and combined with an active contour model and thresholding-based post-processing steps. This method was applied to 20 other MRIs from the OAI and to baseline and follow-up MRIs from a 12-week lower-limb strengthening or endurance training intervention (n = 35 females). The agreement of semi-automated vs. previous manual segmentation was assessed using the Dice similarity coefficient and Bland-Altman analyses. Longitudinal changes observed in the training intervention were compared between semi-automated and manual segmentations.

Results

High agreement was observed between manual and semi-automated segmentations for subcutaneous fat, quadriceps and hamstring CSAs. With strength training, both the semi-automated and manual segmentation method detected a significant reduction in adipose tissue CSA and a significant gain in quadriceps, hamstring and adductor CSAs. With endurance training, a significant reduction in adipose tissue CSAs was observed with both methods.

Conclusion

The semi-automated approach showed high agreement with manual segmentation of thigh muscle and adipose tissue CSAs and showed longitudinal training effects similar to that observed using manual segmentation.

Electronic supplementary material

The online version of this article (doi:10.1007/s10334-017-0622-3) contains supplementary material, which is available to authorized users.

Tissue segmentation: a crucial tool for quantitative MRI and visualization of anatomical structures

Automatic or semi-automatic segmentation of tissue types or organs is well established for X-ray-based computed tomography, with its fixed grey-scale and tissue classes with well-established ranges of Hounsfield units. MRI is much more powerful with regard to soft tissue contrast and quantitative assessment of tissue properties (e.g., perfusion, diffusion, fat content), but the principle of signal generation and recording in MRI leads to inherent problems if simple threshold based segmentation procedures are applied. In this editorial in the special issue of MAGMA on tissue segmentation, a number of relevant methodical, scientific, and clinical aspects of reliable tissue segmentation using data recording by MRI are reported and discussed.