scholarly journals DeepACSA: Automatic segmentation of anatomical cross-sectional area in ultrasound images of human lower limb muscles using deep learning

2021 ◽  
Author(s):  
Paul Ritsche ◽  
Philipp Wirth ◽  
Neil Cronin ◽  
Fabio Sarto ◽  
Marco Narici ◽  
...  
Keyword(s):  
Deep Learning ◽  
Lower Limb ◽  
Vastus Lateralis ◽  
Rectus Femoris ◽  
Cross Sectional Area ◽  
Mean Difference ◽  
Ultrasound Images ◽  
Sectional Area ◽  
Cross Sectional ◽  
Manual Analysis

Background: Muscle anatomical cross-sectional area (ACSA) is an important parameter that characterizes muscle function and helps to classify the severity of several muscular disorders. Ultrasound is a patient friendly, fast and cheap method of assessing muscle ACSA, but manual analysis of the images is laborious, subjective and requires thorough experience. To date, no open access and fully automated program to segment ACSA in ultrasound images is available. On this basis, 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 convolutional neural networks using 1772 ultrasound images from 153 participants (25 females, 128 males; mean age = 38.2 years, range: 13-78) captured by three experienced operators using three distinct devices. We trained three muscle-specific models to detect ACSA. Findings: Comparing DeepACSA analysis of the RF to manual analysis resulted in intra-class correlation (ICC) of 0.96 (95% CI 0.94,0.97), mean difference of 0.31 cm2 (0.04,0.58) and standard error of the differences (SEM) of 0.91 cm2 (0.47,1.36). For the VL, ICC was 0.94 (0.91,0.96), mean difference was 0.25 cm2 (-0.21,0.7) and SEM was 1.55 cm2 (1.13,1.96). The GM/GL muscles demonstrated an ICC of 0.97 (0.95,0.98), a mean difference of 0.01 cm2 (-0.25, 0.24) and a SEM of 0.69 cm2 (0.52,0.83). Interpretation: DeepACSA provides fast and objective segmentation of lower limb panoramic ultrasound images comparable to manual segmentation and is easy to implement both in research and clinical settings. Inaccurate model predictions occurred predominantly on low-quality images, highlighting the importance of high image quality for accurate prediction.

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

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Paul Ritsche ◽  
Philipp Wirth ◽  
Martino V. Franchi ◽  
Oliver Faude
Keyword(s):  
Vastus Lateralis ◽  
Rectus Femoris ◽  
Cross Sectional Area ◽  
Ultrasound Images ◽  
Sectional Area ◽  
Automatic Assessment ◽  
Cross Sectional ◽  
Lower Limb Muscles ◽  
Standard Error Of Measurement ◽  
Error Of Measurement

AbstractOpen-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.

Skeletal Muscle Composition and Glucose Levels in Children Who Are Overweight and Obese

2020 ◽  
Vol 32 (3) ◽  
pp. 157-164
Author(s):  
Trent J. Herda ◽  
Philip M. Gallagher ◽  
Jonathan D. Miller ◽  
Matthew P. Bubak ◽  
Mandy E. Parra
Keyword(s):  
Insulin Resistance ◽  
Skeletal Muscle ◽  
Vastus Lateralis ◽  
Rectus Femoris ◽  
Cross Sectional Area ◽  
First Dorsal Interosseous ◽  
Sectional Area ◽  
Cross Sectional ◽  
Muscle Composition ◽  
Glucose Levels

Background: Skeletal muscle is overlooked in the realm of insulin resistance in children who are overweight and obese despite the fact that it accounts for the most glucose disposal. Objectives: Therefore, this study examined fasted glucose levels and muscle cross-sectional area and echo intensity (EI) via ultrasound images of the first dorsal interosseous, vastus lateralis, and rectus femoris in children who are normal weight and overweight and obese aged 8–10 years. Methods: In total, 13 males (age = 9.0 [0.7] y) and 7 females (age = 9.0 [0.8] y) volunteered for this study. Independent samples t tests and effect sizes (ESs) were used to examine potential differences in skeletal muscle composition and glucose concentrations. Results: There were no significant differences between groups for glucose concentration (P = .07, ES = 0.86); however, the children who were overweight and obese had significantly greater EI (P < .01, ES = 0.98–1.63) for the first dorsal interosseous, vastus lateralis, and rectus femoris and lower cross-sectional area when normalized to EI when collapsed across muscles (P < .04, ES = 0.92). Glucose concentrations correlated with EI and cross-sectional area/EI for the vastus lateralis (r = .514 to −.593) and rectus femoris (r = .551 to −.513), but not the first dorsal interosseous. Discussion: There is evidence that adiposity-related pathways leading to insulin resistance and skeletal muscle degradation are active in young children who are overweight and obese.

scholarly journals Quadriceps cross-sectional area changes in young healthy men with different magnitude of Q angle

2008 ◽  
Vol 105 (3) ◽  
pp. 800-804 ◽  
Author(s):  
Aikaterini E. Tsakoniti ◽  
Christoforos A. Stoupis ◽  
Spyros I. Athanasopoulos
Keyword(s):  
Knee Joint ◽  
Vastus Lateralis ◽  
Dynamic Mri ◽  
Rectus Femoris ◽  
Dynamic Imaging ◽  
Cross Sectional Area ◽  
Protective Mechanism ◽  
Sectional Area ◽  
Cross Sectional ◽  
Q Angle

Knee pain and dysfunction have been often associated with an ineffective pull of the patella by the vastus medialis (VM) relative to the vastus lateralis (VL), particularly in individuals with knee joint malalignment. Such changes in muscular behavior may be attributed to muscle inhibition and/or atrophy that precedes the onset of symptoms. The aim of this study was to investigate possible effects of knee joint malalignment, indicated by a high quadriceps (Q) angle (HQ angle >15°), on the anatomic cross-sectional area (aCSA) of the entire quadriceps and its individual parts, in a group of 17 young asymptomatic men compared with a group of 19 asymptomatic individuals with low Q angle (LQ angle <15°). The aCSA of the entire quadriceps (TQ), VM, VL, vastus intermedius (VI), rectus femoris (RF), and patellar tendon (PT) were measured during static and dynamic magnetic resonance imaging (MRI) with the quadriceps relaxed and under contraction, respectively. A statistically significant lower aCSA was obtained in the HQ angle group, compared with the LQ angle group, for the TQ, VL, and VI in both static (TQ = 9.9%, VL = 12.9%, and VI = 9.1%; P < 0.05) and dynamic imaging (TQ = 10.7%, P < 0.001; VL = 13.4%, P < 0.01; and VI = 9.8%, P < 0.05) and the aCSA of the VM in dynamic MRI (11.9%; P < 0.01). The muscle atrophy obtained in the HQ angle group may be the result of a protective mechanism that inhibits and progressively adapts muscle behavior to reduce abnormal loading and wear of joint structures.

Musicians Have Thicker Median Nerve Cross Sectional Area and More Symptoms of Carpal Tunnel Than Non-Musicians

2020 ◽  
Vol 35 (3) ◽  
pp. 138-144
Author(s):  
Emily Pratt ◽  
Henning Vauth ◽  
Gary McIlvain ◽  
Mark K Timmons
Keyword(s):  
Median Nerve ◽  
Carpal Tunnel ◽  
Hand Function ◽  
Cross Sectional Area ◽  
Mean Difference ◽  
Control Group ◽  
Ultrasound Images ◽  
Sectional Area ◽  
Cross Sectional ◽  
Patient Reported

AIMS: Musicians spend numerous hours perfecting their skills and art, often leading to overuse injuries of the hand; of specific concern to musicians is carpal tunnel syndrome (CTS). This study evaluated the median nerve cross-sectional area and hand function of musicians and made comparisons to non-musicians. METHODS: Seventy-six participants took part in the study, 38 music students and professors in the musician group and 38 participants in the control group. Participants completed patient-reported questionnaires to assess the level of pain and hand function. Ultrasound images were collected at the carpal tunnel inlet and outlet. The median nerve cross-sectional area and the depth of the carpal tunnel were measured on ultrasound images using software imbedded in the ultrasound unit. RESULTS: Musicians showed higher levels of hand dysfunction and CTS symptoms than the non-musicians. The median nerve cross-sectional area was greater in musicians than in the non-musician group on both the right (mean difference 1.5 mm2, p=0.002) and left sides (mean difference 0.9 mm2, p=0.036). The depth of the carpal tunnel at the carpal tunnel inlet and outlet did not differ between the groups (p>0.05). CONCLUSION: The current research identified between-group differences in median nerve cross-sectional area and the level of hand dysfunction. Understanding the interaction between the anatomy of the wrist and wrist and hand dysfunction will benefit clinicians when evaluating and treating musicians.

scholarly journals Intra-rater and inter-rater reliability of the process of obtaining cross-sectional area and echo intensity measurements of muscles from ultrasound images

2021 ◽  
Vol 21 (84) ◽  
pp. e7-e11
Author(s):  
Eric J. Sobolewski ◽  
◽  
Leah D. Wein ◽  
Jacquelyn M. Crow ◽  
Kaitlyn M. Carpenter ◽  
...  
Keyword(s):  
Vastus Lateralis ◽  
Measurement Techniques ◽  
Subcutaneous Fat ◽  
Cross Sectional Area ◽  
Muscle Quality ◽  
Ultrasound Images ◽  
Sectional Area ◽  
Echo Intensity ◽  
Cross Sectional ◽  
Rater Reliability

Introduction: The use of ultrasound images for analyzing muscle quality and size is continuing to grow in the literature. However, many of these manuscripts fail to properly describe their measurement techniques and steps involved in analyzing ultrasound images. Aim of this study: To evaluate the intra- and inter-rater reliability of the steps involved when analyzing ultrasound images to measure cross-sectional area and echo intensity. Material and methods: Twenty ultrasound images of the rectus femoris and vastus lateralis images were blinded and replicated, and then analyzed by experienced raters. The raters then were asked to analyze the images using open-source software for scaling measurements, subcutaneous fat thickness, cross-sectional area, and echo intensity. Matched image values for each measurement where compared for intra- and inter-rater reliability. Results: Intra-rater reliability ranged from fair (ICC3,1 = 0.32) to high (0.98), with echo intensity values being the least reliable (>0.55), and scaling and depth measurements being the most reliable (<0.85). Inter-rater reliability ranged from good (0.77) to high (0.97). Conclusion: Ultrasound- derived measures of cross-sectional area and echo intensity can be measured reliably, with echo intensity being the most difficult to replicate. However, reliability measures are unique to the rater and study and, therefore, should be clearly reported in every paper.

scholarly journals Declines in skeletal muscle quality vs. size following two weeks of knee joint immobilization

PeerJ ◽  
2020 ◽  
Vol 8 ◽  
pp. e8224 ◽  
Author(s):  
Rob J. MacLennan ◽  
Michael Sahebi ◽  
Nathan Becker ◽  
Ethan Davis ◽  
Jeanette M. Garcia ◽  
...  
Keyword(s):  
Skeletal Muscle ◽  
Knee Joint ◽  
Vastus Lateralis ◽  
Rectus Femoris ◽  
Cross Sectional Area ◽  
Muscle Quality ◽  
Muscle Size ◽  
Sectional Area ◽  
Echo Intensity ◽  
Cross Sectional

Background Disuse of a muscle group, which occurs during bedrest, spaceflight, and limb immobilization, results in atrophy. It is unclear, however, if the magnitude of decline in skeletal muscle quality is similar to that for muscle size. The purpose of this study was to examine the effects of two weeks of knee joint immobilization on vastus lateralis and rectus femoris echo intensity and cross-sectional area. Methods Thirteen females (mean ± SD age = 21 ± 2 years) underwent two weeks of left knee joint immobilization via ambulating on crutches and use of a brace. B-mode ultrasonography was utilized to obtain transverse plane images of the immobilized and control vastus lateralis and rectus femoris at pretest and following immobilization. Effect size statistics and two-way repeated measures analyses of variance were used to interpret the data. Results No meaningful changes were demonstrated for the control limb and the rectus femoris of the immobilized limb. Analyses showed a large increase in vastus lateralis echo intensity (i.e., decreased muscle quality) for the immobilized limb (p = .006, Cohen’s d = .918). For vastus lateralis cross-sectional area, no time × limb interaction was observed (p = .103), but the effect size was moderate (d = .570). There was a significant association between the increase in vastus lateralis echo intensity and the decrease in cross-sectional area (r =  − .649, p = .016). Conclusion In female participants, two weeks of knee joint immobilization resulted in greater deterioration of muscle quality than muscle size. Echo intensity appears to be an attractive clinical tool for monitoring muscle quality during disuse.

scholarly journals Rectus femoris muscle thickness and cross-sectional area on ultrasonography may predict isometric and isokinetic knee extension strength: A cross-sectional study

2021 ◽  
Author(s):  
Ufuk Şekir 9) ◽  
Uğur Can Yalaki ◽  
Bedrettin Akova
Keyword(s):  
Vastus Lateralis ◽  
Muscle Thickness ◽  
Rectus Femoris ◽  
Pennation Angle ◽  
Cross Sectional Area ◽  
Isokinetic Strength ◽  
Sectional Area ◽  
Cross Sectional ◽  
Rectus Femoris Muscle ◽  
Femoris Muscle

Objective: To examine the relationship between knee extensor strength and quadriceps muscle architecture evaluated with ultrasonography during relaxed and contracted situations. Materials and Methods: A total of 40 healthy participants (age range 18-40), doing sports at a recreational level were included. Pennation angle, muscle thickness, and cross-sectional area of the vastus medialis, vastus lateralis, and rectus femoris muscles were measured firstly during rest while participants are sitting on an isokinetic dynamometer with their knees at 0° and 60° of flexion. Thereafter, ultrasound evaluations were performed during maximal isometric contraction at 60° knee flexion and maximal isokinetic contraction at 30°/sec and 60°/sec speeds. The architectural parameters were correlated with peak isometric (measured at 60° knee flexion) and isokinetic (measured at 30°/sec and 60°/sec angular velocities) torque values. Results: Pennation angle (p<0.001), muscle thickness (p<0.001) and muscle cross-sectional area (p<0.001) of the vastus medialis muscle during rest, and isometric and isokinetic maximal contractions were higher than the vastus lateralis and rectus femoris muscles. Pennation angle, muscle thickness and muscle cross-sectional area parameters measured during rest, and isometric and isokinetic maximal contractions in the vastus medialis (r=0.39-0.64, p<0.05-0.01) and vastus lateralis (r=0.36-0.68, p<0.05-0.01) showed weak to moderate correlations with isometric and isokinetic peak torque. In rectus femoris muscle, on the other hand, except the weak correlation in pennation angle (r=0.35-0.49, p<0.05-0.01), muscle thickness (r=0.74-0.80, p<0.001) and cross-sectional area (r=0.71-0.80, p<0.001) had a moderate to strong correlation with isometric and isokinetic strength. Stepwise regression analysis indicated that rectus femoris cross-sectional area measured during knee relaxed at 60° flexion (R2=0.532-0.610) and rectus femoris muscle thickness measured during isometric and isokinetic contraction modes (R2=0.538-0.600) were decisive to predict the isometric and isokinetic strength of the quadriceps muscle. Conclusion: Contrary to pennation angle, muscle thickness and cross-sectional area of the rectus femoris measured during relaxed and contracted conditions may be determinative in predicting isometric and isokinetic strength.

scholarly journals Rectus Femoris Mimicking Ultrasound Phantom for Muscle Mass Assessment: Design, Research, and Training Application

2021 ◽  
Vol 10 (12) ◽  
pp. 2721
Author(s):  
Nobuto Nakanishi ◽  
Shigeaki Inoue ◽  
Rie Tsutsumi ◽  
Yusuke Akimoto ◽  
Yuko Ono ◽  
...  
Keyword(s):  
Measurement Accuracy ◽  
Rectus Femoris ◽  
Cross Sectional Area ◽  
Sectional Area ◽  
Cross Sectional ◽  
Phantom Model ◽  
Ultrasound Measurements ◽  
Ultrasound Phantom

Ultrasound has become widely used as a means to measure the rectus femoris muscle in the acute and chronic phases of critical illness. Despite its noninvasiveness and accessibility, its accuracy highly depends on the skills of the technician. However, few ultrasound phantoms for the confirmation of its accuracy or to improve technical skills exist. In this study, the authors created a novel phantom model and used it for investigating the accuracy of measurements and for training. Study 1 investigated how various conditions affect ultrasound measurements such as thickness, cross-sectional area, and echogenicity. Study 2 investigated if the phantom can be used for the training of various health care providers in vitro and in vivo. Study 1 showed that thickness, cross-sectional area, and echogenicity were affected by probe compression strength, probe angle, phantom compression, and varying equipment. Study 2 in vitro showed that using the phantom for training improved the accuracy of the measurements taken within the phantom, and Study 2 in vivo showed the phantom training had a short-term effect on improving the measurement accuracy in a human volunteer. The new ultrasound phantom model revealed that various conditions affected ultrasound measurements, and phantom training improved the measurement accuracy.

Regulation of fiber size, oxidative potential, and capillarization in human muscle by resistance exercise

1999 ◽  
Vol 276 (2) ◽  
pp. R591-R596 ◽  
Author(s):  
H. Green ◽  
C. Goreham ◽  
J. Ouyang ◽  
M. Ball-Burnett ◽  
D. Ranney
Keyword(s):  
Fiber Type ◽  
Vastus Lateralis ◽  
Succinic Dehydrogenase ◽  
Quadriceps Muscle ◽  
Cellular Tissue ◽  
Cross Sectional Area ◽  
Fiber Types ◽  
Sectional Area ◽  
Oxidative Potential ◽  
Cross Sectional

To examine the hypothesis that increases in fiber cross-sectional area mediated by high-resistance training (HRT) would result in a decrease in fiber capillarization and oxidative potential, regardless of fiber type, we studied six untrained males (maximum oxygen consumption, 45.6 ± 2.3 ml ⋅ kg−1 ⋅ min−1; mean ± SE) participating in a 12-wk program designed to produce a progressive hypertrophy of the quadriceps muscle. The training sessions, which were conducted 3 times/wk, consisted of three sets of three exercises, each performed for 6–8 repetitions maximum (RM). Measurements of fiber-type distribution obtained from tissue extracted from the vastus lateralis at 0, 4, 7, and 12 wk indicated reductions ( P < 0.05) in type IIB fibers (15.1 ± 2.1% vs. 7.2 ± 1.3%) by 4 wk in the absence of changes in the other fiber types (types I, IIA, and IIAB). Training culminated in a 17% increase ( P < 0.05) in cross-sectional area by 12 wk with initial increases observed at 4 wk. The increase was independent of fiber type-specific changes. The number of capillaries in contact with each fiber type increased by 12 wk, whereas capillary contacts-to-fiber area ratios remained unchanged. In a defined cross-sectional field, HRT also increased the capillaries per fiber at 12 wk. Training failed to alter cellular oxidative potential, as measured by succinic dehydrogenase (SDH) activity, regardless of fiber type and training duration. It is concluded that modest hypertrophy induced by HRT does not compromise cellular tissue capillarization and oxidative potential regardless of fiber type.

scholarly journals Intra- and inter-muscular differences in the cross-sectional area of the quadriceps muscles assessed by extended field-of-view ultrasonography

2020 ◽  
Vol 22 (2) ◽  
pp. 152 ◽  
Author(s):  
Chrysostomos Sahinis ◽  
Eleftherios Kellis ◽  
Nikiforos Galanis ◽  
Konstantinos Dafkou ◽  
Athanasios Ellinoudis
Keyword(s):  
Vastus Lateralis ◽  
Intraclass Correlation ◽  
Cross Sectional Area ◽  
Field Of View ◽  
Cost Evaluation ◽  
Sectional Area ◽  
Cross Sectional ◽  
Vastus Intermedius ◽  
Extended Field ◽  
Extended Field Of View

Aim: Τo examine the inter- and intra-muscular differences in the anatomical cross-sectional area (CSA) of the quadricep muscles, using extended - field of view (EFOV) ultrasonography (US).Material and methods: Panoramic transverse US images of the thigh were acquired from 10 young participants at five different locations across the thigh, in two sessions, spaced a week apart. The CSA of the vastus medialis (VM), rectus femoris (RF), vastus intermedius (VI), vastus lateralis (VL) and tensor vastus intermedius (TVI) was quantified.Results: The intraclass correlation coefficients ranged from 0.75 to 0.97 and the standard error of measurement ranged from 0.78% to 6.61%, indicating high test-retest reliability. Analysis of the variance indicated that among the 5 quadriceps muscles the VL and the RF displayed the greater CSA proximally, the VI medially and the VM distally across the thigh (p <0.05). No differences in the quadriceps CSA measured with and without including the TVI were found (p >0.05).Conclusions: The EFOV US technique provides transverse scans of the quadriceps muscle in vivo and allowed a reliable and non-invasive determination of CSA at a low cost. Evaluation of CSA along the thigh largely depends on the measurement site. Future studies that examine the quadriceps CSA using EFOV after any form of intervention should consider changes of at least 6.5% as meaningful.

Sign in / Sign up

Export Citation Format

Share Document