Transverse Muscle Ultrasound Analysis (TRAMA): Robust and Accurate Segmentation of Muscle Cross-Sectional Area

Automatic Analysis of Transverse Musculoskeletal Ultrasound Images Based on the Multi-Task Learning Model

Musculoskeletal ultrasound imaging is an important basis for the early screening and accurate treatment of muscle disorders. It allows the observation of muscle status to screen for underlying neuromuscular diseases including myasthenia gravis, myotonic dystrophy, and ankylosing muscular dystrophy. Due to the complexity of skeletal muscle ultrasound image noise, it is a tedious and time-consuming process to analyze. Therefore, we proposed a multi-task learning-based approach to automatically segment and initially diagnose transverse musculoskeletal ultrasound images. The method implements muscle cross-sectional area (CSA) segmentation and abnormal muscle classification by constructing a multi-task model based on multi-scale fusion and attention mechanisms (MMA-Net). The model exploits the correlation between tasks by sharing a part of the shallow network and adding connections to exchange information in the deep network. The multi-scale feature fusion module and attention mechanism were added to MMA-Net to increase the receptive field and enhance the feature extraction ability. Experiments were conducted using a total of 1827 medial gastrocnemius ultrasound images from multiple subjects. Ten percent of the samples were randomly selected for testing, 10% as the validation set, and the remaining 80% as the training set. The results show that the proposed network structure and the added modules are effective. Compared with advanced single-task models and existing analysis methods, our method has a better performance at classification and segmentation. The mean Dice coefficients and IoU of muscle cross-sectional area segmentation were 96.74% and 94.10%, respectively. The accuracy and recall of abnormal muscle classification were 95.60% and 94.96%. The proposed method achieves convenient and accurate analysis of transverse musculoskeletal ultrasound images, which can assist physicians in the diagnosis and treatment of muscle diseases from multiple perspectives.

Muscle Cross-Sectional Area Segmentation in Transverse Ultrasound Images Using Vision Transformers

Automatically measuring a muscle’s cross-sectional area is an important application in clinical practice that has been studied extensively in recent years for its ability to assess muscle architecture. Additionally, an adequately segmented cross-sectional area can be used to estimate the echogenicity of the muscle, another valuable parameter correlated with muscle quality. This study assesses state-of-the-art convolutional neural networks and vision transformers for automating this task in a new, large, and diverse database. This database consists of 2005 transverse ultrasound images from four informative muscles for neuromuscular disorders, recorded from 210 subjects of different ages, pathological conditions, and sexes. Regarding the reported results, all of the evaluated deep learning models have achieved near-to-human-level performance. In particular, the manual vs. the automatic measurements of the cross-sectional area exhibit an average discrepancy of less than 38.15 mm2, a significant result demonstrating the feasibility of automating this task. Moreover, the difference in muscle echogenicity estimated from these two readings is only 0.88, another indicator of the proposed method’s success. Furthermore, Bland−Altman analysis of the measurements exhibits no systematic errors since most differences fall between the 95% limits of agreements and the two readings have a 0.97 Pearson’s correlation coefficient (p < 0.001, validation set) with ICC (2, 1) surpassing 0.97, showing the reliability of this approach. Finally, as a supplementary analysis, the texture of the muscle’s visible cross-sectional area was examined using deep learning to investigate whether a classification between healthy subjects and patients with pathological conditions solely from the muscle texture is possible. Our preliminary results indicate that such a task is feasible, but further and more extensive studies are required for more conclusive results.

DeepACSA: Automatic Segmentation of Cross-Sectional Area in Ultrasound Images of Lower Limb Muscles Using Deep Learning

PURPOSE

Muscle anatomical cross-sectional area (ACSA) can be assessed using ultrasound and images are usually evaluated manually. Here, we present DeepACSA, a deep learning approach to automatically segment ACSA in panoramic ultrasound images of the human rectus femoris (RF), vastus lateralis (VL), gastrocnemius medialis (GM) and lateralis (GL) muscles.

METHODS

We trained three muscle-specific convolutional neural networks (CNN) using 1772 ultrasound images from 153 participants (age = 38.2 yr, range = 13-78). Images were acquired in 10% increments from 30% to 70% of femur length for RF and VL and at 30% and 50% of muscle length for GM and GL. During training, CNN performance was evaluated using intersection-over-union scores. We compared the performance of DeepACSA to manual analysis and a semiautomated algorithm using an unseen test set.

RESULTS

Comparing DeepACSA analysis of the RF to manual analysis with erroneous predictions removed (3.3%) resulted in intraclass correlation (ICC) of 0.989 (95% confidence interval = 0.983-0.992), mean difference of 0.20 cm 2 (0.10-0.30), and SEM of 0.33 cm 2 (0.26-0.41). For the VL, ICC was 0.97 (0.96-0.968), mean difference was 0.85 cm 2 (-0.4 to 1.31), and SEM was 0.92 cm 2 (0.73-1.09) after removal of erroneous predictions (7.7%). After removal of erroneous predictions (12.3%), GM/GL muscles demonstrated an ICC of 0.98 (0.96-0.99), a mean difference of 0.43 cm 2 (0.21-0.65), and an SEM of 0.41 cm 2 (0.29-0.51). Analysis duration was 4.0 ± 0.43 s (mean ± SD) for analysis of one image in our test set using DeepACSA.

CONCLUSIONS

DeepACSA provides fast and objective segmentation of lower limb panoramic ultrasound images comparable with manual segmentation. Inaccurate model predictions occurred predominantly on low-quality images, highlighting the importance of high-quality image for accurate prediction.

Automatic Extraction of Muscle Parameters with Attention UNet in Ultrasonography

Automatically delineating the deep and superficial aponeurosis of the skeletal muscles from ultrasound images is important in many aspects of the clinical routine. In particular, finding muscle parameters, such as thickness, fascicle length or pennation angle, is a time-consuming clinical task requiring both human labour and specialised knowledge. In this study, a multi-step solution for automating these tasks is presented. A process to effortlessly extract the aponeurosis for automatically measuring the muscle thickness has been introduced as a first step. This process consists mainly of three parts. In the first part, the Attention UNet has been incorporated to automatically delineate the boundaries of the studied muscles. Afterwards, a specialised post-processing algorithm was utilised to improve (and correct) the segmentation results. Lastly, the calculation of the muscle thickness was performed. The proposed method has achieved similar to a human-level performance. In particular, the overall discrepancy between the automatic and the manual muscle thickness measurements was equal to 0.4 mm, a significant result that demonstrates the feasibility of automating this task. In the second step of the proposed methodology, the fascicle’s length and pennation angle are extracted through an unsupervised pipeline. Initially, filtering is applied to the ultrasound images to further distinguish the tissues from the other muscle structures. Later, the well-known K-Means algorithm is used to isolate them successfully. As the last step, the dominant angle of the segmented muscle tissues is reported and compared with manual measurements. The proposed pipeline is showing very promising results in the evaluated dataset. Specifically, in the calculation of the pennation angle, the overall discrepancy between the automatic and the manual measurements was less than 2.22° (degrees), once more comparable with the human-level performance. Finally, regarding the fascicle length measurements, the results were divided based on the muscle properties. In the muscles where a large portion (or all) of the fascicles are located between the upper and lower aponeuroses, the proposed pipeline exhibits superb performance; otherwise, overall accuracy deteriorates due to errors caused by the trigonometric approximations needed for the length calculation.

Human skeletal muscle size with ultrasound imaging: a comprehensive review

Skeletal muscle size is an important factor in assessing adaptation to exercise training and detraining, athletic performance, age-associated atrophy and mobility decline, clinical conditions associated with cachexia, and overall skeletal muscle health. Magnetic resonance (MR) imaging and computed tomography (CT) are widely accepted as the gold standard methods for skeletal muscle size quantification. However, it is not always feasible to use these methods (e.g., field studies, bedside studies, large cohort studies). Ultrasound has been available for skeletal muscle examination for more than 50 years and the development, utility, and validity of ultrasound imaging are underappreciated. It is now possible to use ultrasound in situations where MR and CT imaging are not suitable. This review provides a comprehensive summary of ultrasound imaging and human skeletal muscle size assessment. Since the first study in 1968, more than 600 articles have used ultrasound to examine the cross-sectional area and/or volume of 107 different skeletal muscles in more than 27,500 subjects of various ages, health status, and fitness conditions. Data from these studies, supported by decades of technological developments, collectively show that ultrasonography is a valid tool for skeletal muscle size quantification. Considering the wide-ranging connections between human health and function and skeletal muscle mass, the utility of ultrasound imaging will allow it to be employed in research investigations and clinical practice in ways not previously appreciated or considered.

Three-dimensional quantitative muscle ultrasound in a healthy population

INTRODUCTION/AIMS

Quantitative muscle ultrasound offers biomarkers that aid in the diagnosis, detection, and follow-up of neuromuscular disorders. At present, quantitative muscle ultrasound methods are 2D and are often operator and device dependent. The aim of this study was to combine an existing device independent method with an automated ultrasound machine and perform 3D quantitative muscle ultrasound, providing new normative data of healthy controls.

METHODS

In total, 123 healthy volunteers were included. After physical examination, 3D ultrasound scans of the tibialis anterior muscle were acquired using an automated ultrasound scanner. Image postprocessing was performed to obtain calibrated echo intensity values based on a phantom reference.

RESULTS

Tibialis anterior muscle volumes of 61.2 ± 24.1 mL and 53.7 ± 22.7 mL were scanned in males and females, respectively. Echo intensity correlated with gender**, age**, fat fraction*, histogram kurtosis**, skewness* and standard deviation** (*P < .05, **P < .01). Outcome measures did not differ significantly for different acquisition presets. The 3D quantitative muscle ultrasound revealed the non-uniformity of echo intensity values over the length of the tibialis anterior muscle.

DISCUSSION

Our method extended 2D measurements and confirmed previous findings. Our method and reported normative data of (potential) biomarkers can be used to study neuromuscular disorders.

Deep learning segmentation of transverse musculoskeletal ultrasound images for neuromuscular disease assessment

Ultrasound imaging is a patient-friendly and robust technique for studying physiological and pathological muscles. An automatic deep learning (DL) system for the analysis of ultrasound images could be useful to support an expert operator, allowing the study of large datasets requiring less human interaction. The purpose of this study is to present a deep learning algorithm for the cross-sectional area (CSA) segmentation in transverse musculoskeletal ultrasound images, providing a quantitative grayscale analysis which is useful for studying muscles, and to validate the results in a large dataset. The dataset included 3917 images of biceps brachii, tibialis anterior and gastrocnemius medialis acquired on 1283 subjects (mean age 50 ± 21 years, 729 male). The algorithm was based on multiple deep-learning architectures, and its performance was compared to a manual expert segmentation. We compared the mean grayscale value inside the automatic and manual CSA using Bland-Altman plots and a correlation analysis. Classification in healthy and abnormal muscles between automatic and manual segmentation were compared using the grayscale value z-scores. In the test set, a Precision of 0.88 ± 0.12 and a Recall of 0.92 ± 0.09 was achieved. The network segmentation performance was slightly less in abnormal muscles, without a loss of discrimination between healthy and abnormal muscle images. Bland-Altman plots showed no clear trend in the error distribution and the two readings have a 0.99 Pearson's correlation coefficient (p < 0.001, test set). The ICC(A, 1) calculated between the z-score readings was 0.99. The algorithm achieves robust CSA segmentation performance and gives mean grayscale level information comparable to a manual operator. This could provide a helpful tool for clinicians in neuromuscular disease diagnosis and follow-up. The entire dataset and code are made available for the research community.

ACSAuto-semi-automatic assessment of human vastus lateralis and rectus femoris cross-sectional area in ultrasound images

Open-access scripts to perform muscle anatomical cross-sectional area (ACSA) evaluation in ultrasound images are currently unavailable. This study presents a novel semi-automatic ImageJ script (named "ACSAuto") for quantifying the ACSA of lower limb muscles. We compared manual ACSA measurements from 180 ultrasound scans of vastus lateralis (VL) and rectus femoris (RF) muscles to measurements assessed by the ACSAuto script. We investigated inter- and intra-investigator reliability of the script. Consecutive-pairwise intra-class correlations (ICC) and standard error of measurement (SEM) with 95% compatibility interval were calculated. Bland-Altman analyses were employed to test the agreement between measurements. Comparing manual and ACSAuto measurements, ICCs and SEMs ranged from 0.96 to 0.999 and 0.12 to 0.96 cm2 (1.2-5.9%) and mean bias was smaller than 0.5 cm2 (4.3%). Inter-investigator comparison revealed ICCs, SEMs and mean bias ranging from 0.85 to 0.999, 0.07 to 1.16 cm2 (0.9-7.6%) and - 0.16 to 0.66 cm2 (- 0.6 to 3.2%). Intra-investigator comparison revealed ICCs, SEMs and mean bias between 0.883-0.998, 0.07-0.93 cm2 (1.1-7.6%) and - 0.80 to 0.15 cm2 (- 3.4 to 1.8%). Image quality needs to be high for efficient and accurate ACSAuto analyses. Taken together, the ACSAuto script represents a reliable tool to measure RF and VL ACSA, is comparable to manual analysis and can reduce time needed to evaluate ultrasound images.

Assessment methods of vastus lateralis muscle architecture using panoramic ultrasound: a new approach, test-retest reliability and measurement error

Abstract Extended-field-of-view ultrasonography is a valid alternative to determine the dimensions of the skeletal striated muscle; however, some factors may influence the final measurement. The aim of this study was to determine the test-retest reliability and measurement error of vastus lateralis muscle architecture variables through internal anatomical landmarks and to compare three fixed determined points using extended-field-of-view ultrasonography. Twelve young (24 ± 6 years) adult university male students participated in the study. Images were obtained through extended-field-of-view ultrasonography of the vastus lateralis muscle. Measurements were made for muscle thickness (MT), fascicle length (FL), and fascicle pennation angle (FA) using a method that identifies internal anatomical landmarks. MT was also measured at predetermined distances of 2 cm proximal, 6 cm proximal, and 2 cm distal. One-way ANOVA with repeated measures did not identify any test-retest significant differences for all variables measured. Typical measurement error in centimeters (cm) or degrees (º), coefficient of variation in percentage (%) and intraclass correlation coefficient were MT = 0.07 cm, 2.93%, 0.964; FL = 0.31 cm, 2.89%, 0.947; FA = 0.92°, 4.08%, 0.942; MT 2 cm proximal = 0.10 cm, 3.77%, 0.910; MT 6 cm proximal = 0.27 cm, 9.66%, 0.576; MT 2 cm distal = 0.35 cm, 19.76%, 0.564. MT, FL and FA showed high reliability and low measurement error. Internal anatomical landmarks proved to be more reliable and presented smaller measurement errors when compared to the predetermined distances method.

LUMINOUS database: lumbar multifidus muscle segmentation from ultrasound images

BACKGROUND

Among the paraspinal muscles, the structure and function of the lumbar multifidus (LM) has become of great interest to researchers and clinicians involved in lower back pain and muscle rehabilitation. Ultrasound (US) imaging of the LM muscle is a useful clinical tool which can be used in the assessment of muscle morphology and function. US is widely used due to its portability, cost-effectiveness, and ease-of-use. In order to assess muscle function, quantitative information of the LM must be extracted from the US image by means of manual segmentation. However, manual segmentation requires a higher level of training and experience and is characterized by a level of difficulty and subjectivity associated with image interpretation. Thus, the development of automated segmentation methods is warranted and would strongly benefit clinicians and researchers. The aim of this study is to provide a database which will contribute to the development of automated segmentation algorithms of the LM.

CONSTRUCTION AND CONTENT

This database provides the US ground truth of the left and right LM muscles at the L5 level (in prone and standing positions) of 109 young athletic adults involved in Concordia University's varsity teams. The LUMINOUS database contains the US images with their corresponding manually segmented binary masks, serving as the ground truth. The purpose of the database is to enable development and validation of deep learning algorithms used for automatic segmentation tasks related to the assessment of the LM cross-sectional area (CSA) and echo intensity (EI). The LUMINOUS database is publicly available at http://data.sonography.ai .

CONCLUSION

The development of automated segmentation algorithms based on this database will promote the standardization of LM measurements and facilitate comparison among studies. Moreover, it can accelerate the clinical implementation of quantitative muscle assessment in clinical and research settings.

Automatic segmentation of ultrasound images of gastrocnemius medialis with different echogenicity levels using convolutional neural networks

The purpose of this study was to develop an automatic method for the segmentation of muscle cross-sectional area on transverse B-mode ultrasound images of gastrocnemius medialis using a convolutional neural network(CNN). In the provided dataset images with both normal and increased echogenicity are present. The manually annotated dataset consisted of 591 images, from 200 subjects, 400 relative to subjects with normal echogenicity and 191 to subjects with augmented echogenicity. From the DICOM files, the image has been extracted and processed using the CNN, then the output has been post-processed to obtain a finer segmentation. Final results have been compared to the manual segmentations. Precision and Recall scores as mean ± standard deviation for training, validation, and test sets are 0.96 ± 0.05, 0.90 ± 0.18, 0.89 ± 0.15 and 0.97 ±0.03, 0.89± 0.17, 0.90 ± 0.14 respectively. The CNN approach has also been compared to another automatic algorithm, showing better performances. The proposed automatic method provides an accurate estimation of muscle cross-sectional area in muscles with different echogenicity levels.

Utility of Plasma GDF-15 for Diagnosis and Prognosis Assessment of ICU-Acquired Weakness in Mechanically Ventilated Patients: Prospective Observational Study

Objective

To identify the clinical correlations between plasma growth differentiation factor-15 (GDF-15), skeletal muscle function, and acute muscle wasting in ICU patients with mechanical ventilation. In addition, to investigate its diagnostic value for ICU-acquired weakness (ICU-AW) and its predictive value for 90-day survival in mechanically ventilated patients.

Methods

95 patients with acute respiratory failure, who required mechanical ventilation therapy, were randomly selected among hospitalized patients from June 2017 to January 2019. The plasma GDF-15 level was detected by ELISA, the rectus femoris cross-sectional area (RFcsa) was measured by ultrasound, and the patient's muscle strength was assessed using the British Medical Research Council (MRC) muscle strength score on day 1, day 4, and day 7. Patients were divided into an ICU-AW group and a non-ICU-AW group according to their MRC-score on the 7th day. The differences in plasma GDF-15 level, MRC-score, and RFcsa between the two groups were compared on the 1st, 4th, and 7th day after being admitted to the ICU. Then, the correlations between plasma GDF-15 level, RFcsa loss, and MRC-score on day 7 were investigated. The receiver operating characteristic curve (ROC) was used to analyze the plasma GDF-15 level, RFcsa loss, and % decrease in RFcsa on the 7th day to the diagnosis of ICU-AW in mechanically ventilated patients. Moreover, the predictive value of GDF-15 on the 90-day survival status of patients was assessed using patient survival curves.

Results

Based on whether the 7th day MRC-score was <48, 50 cases were included in the ICU-AW group and 45 cases in the non-ICU-AW group. The length of mechanical ventilation, ICU length of stay, and hospital length of stay were significantly longer in the ICU-AW group than in the non-ICU-AW group (all P < 0.05), while the other baseline indicators were not statistically significant between the two groups. As the treatment time increased, the plasma GDF-15 level was significantly increased, the ICU-AW group demonstrated a significant decreasing trend in the MRC-score and RFcsa, while no significant changes were found in the non-ICU-AW group. In the ICU-AW group, the plasma GDF-15 level was significantly higher than that in the non-ICU-AW group, while the RFcsa and the MRC-score were significantly lower than those in the non-ICU-AW group (GDF-15 (pg/ml): 2542.44 ± 629.38 vs. 1542.86 ± 502.86; RFcsa (cm²): 2.04 ± 0.64 vs. 2.34 ± 0.61; MRC-score: 41.22 ± 3.42 vs. 51.42 ± 2.72, all P < 0.05), while the other baseline indicators were not statistically significant between the two groups. As the treatment time increased, the plasma GDF-15 level was significantly increased, the ICU-AW group demonstrated a significant decreasing trend in the MRC-score and RFcsa, while no significant changes were found in the non-ICU-AW group. In the ICU-AW group, the plasma GDF-15 level was significantly higher than that in the non-ICU-AW group, while the RFcsa and the MRC-score were significantly lower than those in the non-ICU-AW group (GDF-15 (pg/ml): 2542.44 ± 629.38 vs. 1542.86 ± 502.86; RFcsa (cm²): 2.04 ± 0.64 vs. 2.34 ± 0.61; MRC-score: 41.22 ± 3.42 vs. 51.42 ± 2.72, all r = −0.60), while it was significantly positively correlated with the RFcsa loss (r = −0.60), while it was significantly positively correlated with the RFcsa loss (r = −0.60), while it was significantly positively correlated with the RFcsa loss (r = −0.60), while it was significantly positively correlated with the RFcsa loss (P < 0.05), while the other baseline indicators were not statistically significant between the two groups. As the treatment time increased, the plasma GDF-15 level was significantly increased, the ICU-AW group demonstrated a significant decreasing trend in the MRC-score and RFcsa, while no significant changes were found in the non-ICU-AW group. In the ICU-AW group, the plasma GDF-15 level was significantly higher than that in the non-ICU-AW group, while the RFcsa and the MRC-score were significantly lower than those in the non-ICU-AW group (GDF-15 (pg/ml): 2542.44 ± 629.38 vs. 1542.86 ± 502.86; RFcsa (cm²): 2.04 ± 0.64 vs. 2.34 ± 0.61; MRC-score: 41.22 ± 3.42 vs. 51.42 ± 2.72, all P < 0.05), while the other baseline indicators were not statistically significant between the two groups. As the treatment time increased, the plasma GDF-15 level was significantly increased, the ICU-AW group demonstrated a significant decreasing trend in the MRC-score and RFcsa, while no significant changes were found in the non-ICU-AW group. In the ICU-AW group, the plasma GDF-15 level was significantly higher than that in the non-ICU-AW group, while the RFcsa and the MRC-score were significantly lower than those in the non-ICU-AW group (GDF-15 (pg/ml): 2542.44 ± 629.38 vs. 1542.86 ± 502.86; RFcsa (cm²): 2.04 ± 0.64 vs. 2.34 ± 0.61; MRC-score: 41.22 ± 3.42 vs. 51.42 ± 2.72, all P < 0.05), while the other baseline indicators were not statistically significant between the two groups. As the treatment time increased, the plasma GDF-15 level was significantly increased, the ICU-AW group demonstrated a significant decreasing trend in the MRC-score and RFcsa, while no significant changes were found in the non-ICU-AW group. In the ICU-AW group, the plasma GDF-15 level was significantly higher than that in the non-ICU-AW group, while the RFcsa and the MRC-score were significantly lower than those in the non-ICU-AW group (GDF-15 (pg/ml): 2542.44 ± 629.38 vs. 1542.86 ± 502.86; RFcsa (cm²): 2.04 ± 0.64 vs. 2.34 ± 0.61; MRC-score: 41.22 ± 3.42 vs. 51.42 ± 2.72, all

Conclusion

The plasma GDF-15 concentration level was significantly associated with skeletal muscle function and muscle wasting on day 7 in ICU patients with mechanical ventilation. Therefore, it can be concluded that the plasma GDF-15 level on the 7th day has a high diagnostic yield for ICU-acquired muscle weakness, and it can predict the 90-day survival status of ICU mechanically ventilated patients.