Segmentation of the fascia lata and reproducible quantification of intermuscular adipose tissue (IMAT) of the thigh

Oliver Chaudry, Andreas Friedberger, Alexandra Grimm, Michael Uder, Armin Michael Nagel, Wolfgang Kemmler, Klaus Engelke, Oliver Chaudry, Andreas Friedberger, Alexandra Grimm, Michael Uder, Armin Michael Nagel, Wolfgang Kemmler, Klaus Engelke

Abstract

Objective: To develop a precise semi-automated segmentation of the fascia lata (FL) of the thigh to quantify IMAT volume in T1w MR images and fat fraction (FF) in Dixon MR images.

Materials and methods: A multi-step segmentation approach was developed to identify fibrous structures of the FL and combining them into a closed 3D surface. 23 healthy young men with low and 50 elderly sarcopenic men with moderate levels of IMAT were measured by T1w and 6pt Dixon MRI at 3T. 20 datasets were used to determine reanalysis precision errors. IMAT volume was compared using the new FL segmentation versus an easier to segment but less accurate, tightly fitting envelope of the thigh muscle ensemble.

Results: The segmentation was successfully applied to all 73 datasets and took about 7 min per 28 slices. In particular, in elderly subjects, it includes a large amount of adipose tissue below the FL typically not accounted for in other segmentation approaches. Inter- and intra-operator RMS-CVs were 0.33% and 0.14%, respectively, for IMAT volume and 0.04% and 0.02%, respectively, for FFMT.

Discussion: The FL segmentation is an important step to quantify IMAT with high precision and may be useful to investigate effects of aging and treatment on changes of IMAT and FF. ClinicalTrials.gov identifier NCT2857660, August 5, 2016.

Trial registration: ClinicalTrials.gov identifier NCT2857660, August 5, 2016.

Trial registration: ClinicalTrials.gov NCT02857660.

Keywords: Adipose tissue; Fascia lata; MRI.

Conflict of interest statement

The authors declare that they have no conflict of interest.

Figures

Fig. 1
Fig. 1
Examples of bias field corrected T1w images of the two subject groups. Elderly sarcopenic subject of group G2 (a); a healthy young subject of group G1 (b). The green contour indicates the position of the fascia lata (FL), the red contour indicates the fascia of the vena saphena magna (also marked in red). These two fasciae are usually highly visible but have to be distinguished
Fig. 2
Fig. 2
a Flowchart of the segmentation process. b Bias field corrected T1w image, which is the input for the fuzzy c-means clustering. c Result of step 1: fibrous tissue (FT, red), bone and background (BG, blue), muscle (green) and adipose tissue (AT, yellow) clusters. Sometimes veins and thicker connective tissue can also lie within the bone and BG cluster. d Result of step 2: tightly fitting muscle envelope
Fig. 3
Fig. 3
Schematic presentation of the filtering process to obtain relevant FL structures. Fibrous tissue cluster obtained from fuzzy clustering (left). Result after filtering all fibrous structures outside the muscle envelope (right). Green: 3D plate-like structures, probably part of the FL; red: structures unlikely part of the FL; for visualization only two colors, green and red are used. In reality, all voxels containing fibrous structures were continuously scaled between 0 and 1 (see text). Grey: region defined by tight muscle envelope
Fig. 4
Fig. 4
Incorrect FL segmentation in red (a). Application of an intelligent scissors tool using manually set seed points (yellow crosses); corrected fascia segmentation (b)
Fig. 5
Fig. 5
Separation of adipose and muscle tissue within the intra-fascia VOI uisng Dixon 6pt images. a The histograms show the grey value distribution (FF values ranging from 0 to 1000) of the whole intra-fascia VOI; the normal histogram (black) and the logarithmically scaled histogram (color gradient). The color gradient indicates muscle tissue (red) and adipose tissue (yellow). The minimum of this distribution (blue line) was used as a threshold to segment muscle tissue. b Segmented Dixon images; red: borders of muscle tissue (black); green: fascia lata; blue: outer surface of the thigh; the grey voxels within the FL denote IMAT
Fig. 6
Fig. 6
Comparison of IMAT results. a Correlation of IMAT volume determined by the FL (IMATFL) versus the narrow muscle envelope segmentation (IMATME). b Absolute IMATFL values plotted against the difference of IMATFL and IMATME. Green dots represent young and blue triangles elderly subjects. One outlier is marked by a black circle. An image of this subject is shown as inset
Fig. 7
Fig. 7
Segmentation of the T1w images of a healthy young (a) and an elderly (b) subject. Image (c) shows the difference in IMAT (magenta) using the segmentation of the FL versus a tight fitting envelope. d A 3D projection of the FL segmentation in green, with muscle in red and AT in yellow

References

    1. Cohen S, Nathan JA, Goldberg AL. Muscle wasting in disease: molecular mechanisms and promising therapies. Nat Rev Drug Discov. 2015;14(1):58–74. doi: 10.1038/nrd4467.
    1. Addison O, Marcus RL, Lastayo PC, Ryan AS. Intermuscular fat: a review of the consequences and causes. Int J Endocrinol. 2014;2014:309570. doi: 10.1155/2014/309570.
    1. McGregor RA, Cameron-Smith D, Poppitt SD. It is not just muscle mass: a review of muscle quality, composition and metabolism during ageing as determinants of muscle function and mobility in later life. Longev Healthspan. 2014;3(1):9. doi: 10.1186/2046-2395-3-9.
    1. Engelke K, Museyko O, Wang L, Laredo JD. Quantitative analysis of skeletal muscle by computed tomography imaging-State of the art. J Orthop Translat. 2018;15:91–103. doi: 10.1016/j.jot.2018.10.004.
    1. Eggers H, Bornert P. Chemical shift encoding-based water-fat separation methods. J Magn Reson Imaging. 2014;40(2):251–268. doi: 10.1002/jmri.24568.
    1. Damon BM, Li K, Bryant ND. Magnetic resonance imaging of skeletal muscle disease. Handb Clin Neurol. 2016;136:827–842. doi: 10.1016/B978-0-444-53486-6.00041-7.
    1. Burakiewicz J, Sinclair CDJ, Fischer D, Walter GA, Kan HE, Hollingsworth KG. Quantifying fat replacement of muscle by quantitative MRI in muscular dystrophy. J Neurol. 2017;264(10):2053–2067. doi: 10.1007/s00415-017-8547-3.
    1. Ten Dam L, van der Kooi AJ, Verhamme C, Wattjes MP, de Visser M. Muscle imaging in inherited and acquired muscle diseases. Eur J Neurol. 2016;23(4):688–703. doi: 10.1111/ene.12984.
    1. Strijkers GJ, Araujo ECA, Azzabou N, Bendahan D, Blamire A, Burakiewicz J, Carlier PG, Damon B, Deligianni X, Froeling M, Heerschap A, Hollingsworth KG, Hooijmans MT, Karampinos DC, Loudos G, Madelin G, Marty B, Nagel AM, Nederveen AJ, Nelissen JL, Santini F, Scheidegger O, Schick F, Sinclair C, Sinkus R, de Sousa PL, Straub V, Walter G, Kan HE. 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. J Neuromuscul Dis. 2019;6(1):1–30. doi: 10.3233/JND-180333.
    1. Mercuri E, Talim B, Moghadaszadeh B, Petit N, Brockington M, Counsell S, Guicheney P, Muntoni F, Merlini L. Clinical and imaging findings in six cases of congenital muscular dystrophy with rigid spine syndrome linked to chromosome 1p (RSMD1) Neuromuscul Disord. 2002;12(7–8):631–638. doi: 10.1016/S0960-8966(02)00023-8.
    1. de Carvalho Felinto J, Poloni KM, de Lima Freire PG, Aily JB, de Almeida AC, Pedroso MG, Mattiello SM, Ferrari RJ. Computational Science and Its Applications—ICCSA 2018. Cham: Springer International Publishing; 2018. Automatic segmentation and quantification of thigh tissues in CT images; pp. 261–276.
    1. Muhlberg A, Museyko O, Laredo JD, Engelke K. A reproducible semi-automatic method to quantify the muscle-lipid distribution in clinical 3D CT images of the thigh. PLoS One. 2017;12(4):e0175174. doi: 10.1371/journal.pone.0175174.
    1. Snyder WS, Cook MJ, Nasset ES, Karhausen LR, Howells GP, Tipton IH. Report of the task group on reference man. Oxford, UK: Pergamon Press; 1975.
    1. Wronska A, Kmiec Z. Structural and biochemical characteristics of various white adipose tissue depots. Acta Physiol (Oxf) 2012;205(2):194–208. doi: 10.1111/j.1748-1716.2012.02409.x.
    1. Karampinos DC, Baum T, Nardo L, Alizai H, Yu H, Carballido-Gamio J, Yap SP, Shimakawa A, Link TM, Majumdar S. Characterization of the regional distribution of skeletal muscle adipose tissue in type 2 diabetes using chemical shift-based water/fat separation. J Magn Reson Imaging. 2012;35(4):899–907. doi: 10.1002/jmri.23512.
    1. Grimm A, Meyer H, Nickel MD, Nittka M, Raithel E, Chaudry O, Friedberger A, Uder M, Kemmler W, Engelke K, Quick HH. A comparison between 6-point dixon MRI and MR spectroscopy to quantify muscle fat in the thigh of subjects with sarcopenia. J Frailty Aging. 2019;8(1):21–26.
    1. Graffy PM, Liu J, Pickhardt PJ, Burns JE, Yao J, Summers RM. Deep learning-based muscle segmentation and quantification at abdominal CT: application to a longitudinal adult screening cohort for sarcopenia assessment. Br J Radiol. 2019;92(1100):20190327. doi: 10.1259/bjr.20190327.
    1. Positano V, Christiansen T, Santarelli MF, Ringgaard S, Landini L, Gastaldelli A. Accurate segmentation of subcutaneous and intermuscular adipose tissue from MR images of the thigh. J Magn Reson Imaging. 2009;29(3):677–684. doi: 10.1002/jmri.21699.
    1. Orgiu S, Lafortuna CL, Rastelli F, Cadioli M, Falini A, Rizzo G. Automatic muscle and fat segmentation in the thigh from T1-Weighted MRI. J Magn Reson Imaging. 2016;43(3):601–610. doi: 10.1002/jmri.25031.
    1. Lareau-Trudel E, Le Troter A, Ghattas B, Pouget J, Attarian S, Bendahan D, Salort-Campana E. Muscle quantitative MR imaging and clustering analysis in patients with facioscapulohumeral muscular dystrophy type 1. PLoS One. 2015;10(7):e0132717. doi: 10.1371/journal.pone.0132717.
    1. Visser M, Goodpaster BH, Kritchevsky SB, Newman AB, Nevitt M, Rubin SM, Simonsick EM, Harris TB. Muscle mass, muscle strength, and muscle fat infiltration as predictors of incident mobility limitations in well-functioning older persons. J Gerontol A Biol Sci Med Sci. 2005;60(3):324–333. doi: 10.1093/gerona/60.3.324.
    1. Marcus RL, Addison O, Kidde JP, Dibble LE, Lastayo PC. Skeletal muscle fat infiltration: impact of age, inactivity, and exercise. J Nutr Health Aging. 2010;14(5):362–366. doi: 10.1007/s12603-010-0081-2.
    1. Kemmler W, Weissenfels A, Teschler M, Willert S, Bebenek M, Shojaa M, Kohl M, Freiberger E, Sieber C, Sv Stengel. Whole-body electromyostimulation and protein supplementation favorably affect sarcopenic obesity in community-dwelling older men at risk: the randomized controlled FranSO study. Clin Interv Aging. 2017;12:1503–1513. doi: 10.2147/CIA.S137987.
    1. Grimm A, Nickel MD, Chaudry O, Uder M, Jakob F, Kemmler W, Quick HH, Engelke K. Feasibility of Dixon magnetic resonance imaging to quantify effects of physical training on muscle composition—a pilot study in young and healthy men. Eur J Radiol. 2019;114:160–166. doi: 10.1016/j.ejrad.2019.03.019.
    1. Grimm A, Meyer H, Nickel MD, Nittka M, Raithel E, Chaudry O, Friedberger A, Uder M, Kemmler W, Engelke K, Quick HH. Repeatability of Dixon magnetic resonance imaging and magnetic resonance spectroscopy for quantitative muscle fat assessments in the thigh. J Cachexia Sarcopenia Muscle. 2018;9(6):1093–1100. doi: 10.1002/jcsm.12343.
    1. Caggiati A. Fascial relationships of the long saphenous vein. Circulation. 1999;100(25):2547–2549. doi: 10.1161/01.CIR.100.25.2547.
    1. Tustison NJ, Avants BB, Cook PA, Zheng Y, Egan A, Yushkevich PA, Gee JC. N4ITK: improved N3 bias correction. IEEE Trans Med Imaging. 2010;29(6):1310–1320. doi: 10.1109/TMI.2010.2046908.
    1. Caselles V, Kimmel R, Sapiro G. Geodesic active contours. Int J Comput Vision. 1997;22(1):61–79. doi: 10.1023/A:1007979827043.
    1. Frangi AF, Niessen WJ, Vincken KL, Viergever MA. Medical image computing and computer-assisted intervention—MICCAI’98. Berlin: Springer; 1998. Multiscale vessel enhancement filtering; pp. 130–137.
    1. Mortensen EN, Barrett WA (1995) Intelligent scissors for image composition. Paper presented at the Proceedings of the 22nd annual conference on Computer graphics and interactive techniques
    1. Hart P, Nilsson N, Raphael B. A formal basis for the heuristic determination of minimum cost paths. IEEE Trans Syst Sci Cybern. 1968;4(2):100–107. doi: 10.1109/TSSC.1968.300136.
    1. R Core Team (2019) R: a language and environment for statistical computing. 3.6.2 edn. R Foundation for Statistical Computing
    1. Glüer C-C, Blake G, Lu Y, Blunt BA, Jergas M, Genant HK. Accurate assessment of precision errors: how to measure the reproducibility of bone densitometry techniques. Osteoporos Int. 1995;5(4):262–270. doi: 10.1007/BF01774016.
    1. Gadermayr M, Li K, Muller M, Truhn D, Kramer N, Merhof D, Gess B. Domain-specific data augmentation for segmenting MR images of fatty infiltrated human thighs with neural networks. J Magn Reson Imaging. 2019;49(6):1676–1683. doi: 10.1002/jmri.26544.
    1. Kovacs W, Liu C-Y, Summers R, Yao J. Identification of muscle and subcutaneous and intermuscular adipose tissue on thigh MRI of muscular dystrophy. 2016 doi: 10.1109/isbi.2016.7493238:176-179.
    1. Gadermayr M, Disch C, Muller M, Merhof D, Gess B. A comprehensive study on automated muscle segmentation for assessing fat infiltration in neuromuscular diseases. Magn Reson Imaging. 2018;48:20–26. doi: 10.1016/j.mri.2017.12.014.
    1. Ronneberger O, Fischer P, Brox T. Medical image computing and computer-assisted intervention—MICCAI 2015. Cham: Springer; 2015. U-Net: convolutional networks for biomedical image segmentation; pp. 234–241.

Source: PubMed

Patrocinadores y Colaboradores

Condiciones médicas

Intervenciones de drogas

3
Suscribir