MuViSS : Muscle, Visceral and Subcutaneous Segmentation by an automatic evaluation method using Deep Learning

MuViSS : Muscle, Visceral and Subcutaneous Segmentation by an automatic evaluation method using Deep Learning | medRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-P4HH5NV'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search MuViSS : Muscle, Visceral and Subcutaneous Segmentation by an automatic evaluation method using Deep Learning Edouard Wasielewski , View ORCID Profile Karim Boudjema , View ORCID Profile Laurent Sulpice , View ORCID Profile Thierry Pecot doi: https://doi.org/10.1101/2024.03.11.24304074 Edouard Wasielewski 1 Department of Hepatobiliary and Digestive Surgery, University Hospital, Rennes 1 University , Rennes, France 2 University of Rennes 1 , Rennes, France 3 INSERM, CIC1414 Centre d’Investigation Clinique de Rennes , Rennes, France Find this author on Google Scholar Find this author on PubMed Search for this author on this site Karim Boudjema 1 Department of Hepatobiliary and Digestive Surgery, University Hospital, Rennes 1 University , Rennes, France 2 University of Rennes 1 , Rennes, France Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Karim Boudjema Laurent Sulpice 1 Department of Hepatobiliary and Digestive Surgery, University Hospital, Rennes 1 University , Rennes, France 2 University of Rennes 1 , Rennes, France 3 INSERM, CIC1414 Centre d’Investigation Clinique de Rennes , Rennes, France Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Laurent Sulpice Thierry Pecot 2 University of Rennes 1 , Rennes, France 4 Univ Rennes, Biosit UAR 3480 CNRS – US 018 Inserm, FAIIA core facility , F-35000 Rennes, France Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Thierry Pecot For correspondence: Thierry.pecot{at}univ-rennes.fr Abstract Full Text Info/History Metrics Data/Code Preview PDF Abstract Purpose Patient body composition is a major factor in patient management. Indeed, assessment of SMI as well as VFA and, to a lesser extent, SFA is a major factor in patient survival, particularly in surgery. However, to date, there is no simple, rapid, open-access assessment method. The aim of this work is to provide a simple, rapid and accurate tool for assessing patients’ body composition. Material and methods A total of 343 patients underwent liver transplantation at the University Hospital of Rennes between January 1 st , 2012 and December 31 s , 2018. Image analysis was performed using the open source software ImageJ. Tissue distinction was based on Hounsfield density. The training dataset used 332 images (320 for training and 12 for validation). The model was evaluated on 11 patients. The complete software and video package is available at https://github.com/tpecot/MuViSS . Results In total, the model was trained with 332 images and evaluated on 11 images. Model accuracy is 0.974 (SD 0.003), Jaccard’s index is 0.98 for visceral fat, 0.895 for muscle and 0.94 for subcutaneous fat. The Dice index is 0.958 (SD 0.003) for visceral fat, 0.944 (SD: 0.012) for muscle and 0.970 (SD: 0.013) for subcutaneous fat. Finally, the Normalized root mean square error is 0.007 for visceral fat, 0.0518 for muscle and 0.0124 for subcutaneous fat. Conclusion To our knowledge, this is the first freely available model for assessing body composition. The model is fast, simple and accurate, based on Deep Learning. Statements and declarations All authors declare no conflict of interest INTRODUCTION Sarcopenia is defined by the European Working Group on Sarcopenia in Older People (EWGSOP) as the progressive and generalized reduction of skeletal muscle mass. Well known in the elderly, its incidence is also high in many chronic diseases such as respiratory [ 1 ], renal [ 2 ] or liver diseases [ 3 ]. Sarcopenia, evaluated by the Sketal Muscule Index (SMI) (ratio between the area of the muscles at L3 level and the height squared) is found to be a risk factor for poor prognosis in surgery, in particular with a decrease in the overall survival of patients undergoing hepatic [ 4 ], colorectal [ 5 ] or pancreatic surgery [ 6 ]. Obesity, a true epidemic on a global scale, is defined by the WHO as an abnormal or excessive accumulation of fat, representing a real health risk. Indeed, 4 million people die each year as a result of being overweight or obese. In addition to the risk of developing chronic diseases such as coronary artery disease [ 7 ], hypertension [ 8 ], dyslipidemia [ 9 ] or NASH [ 10 ], obesity is recognized in cancerology as a risk factor for many cancers [ 11 ], including 13 with a high level of evidence [ 12 ]. Moreover, obesity represents a risk factor for mortality in this population [ 13 ]. Although global obesity is strongly correlated with abdominal obesity, these 2 parameters are not always linked. In fact, there are patients who are obese, based on BMI calculation, but not viscerally obese, and vice versa. However, this distinction is relatively difficult to assess in clinical practice [ 14 ]. The assessment of body composition is complex. Indeed, the gold standard remains the biophotonic X-ray absorptiometry. However, the rarity, the cost and the method (prospective evaluation) do not allow an evaluation in current practice. Computed Tomography (CT), which is faster and widely used for diagnosis and follow-up, is the tool of choice for retrospective assessment of body composition. Several articles in the literature showed that there is a strong correlation between CT data and body composition [ 15 – 18 ]. Therefore, sarcopenia is assessed by the skeletal muscle index (SMI), division of SMA (skeletal muscle mass) by size squared, on the third lumbar vertebra [ 19 , 20 ]. According to the recommendations [ 21 ], the assessment of abdominal obesity involves the measurement of the VFA (Visceral Fat Area) on a CT scan. Although widely used, there are few freely available tools using DeepLearning to assess SMA (skeletal muscle area), VFA (visceral fat area) and SFA (subcutaneous fat area) simply and quickly. In this study, we provide an automated method for assessing SMA, VFA and SFA at the L3 level. MATERIAL AND METHODS Patient A total of 331 patients underwent liver transplantation at the University Hospital of Rennes between January 1 st , 2012 and December 31 s , 2018. The evaluation of the SMA (Skeletal Muscul Area), SFA (Subcutaneous Fat Area) and VFA (Visceral Fat Area) were assessed from the last CT scan performed in the 90 days before transplantation. The data used to build the MuViss model comes from another study, the results of which have not yet been published. However, the study in question, from which the data are taken, has been approved by the ethics committee of the University Hospital of Rennes. Image Analysis Analysis was performed on a slice through the third lumbar vertebra. The images were processed using ImageJ version 2.3.0/1.53f (open source) (no image analysis software having demonstrated superiority). The distinction between different tissues was based on the Hounsfield Unit (HU). The threshold of -29 to + 150 HU was used for skeletal muscle, which at L3 included: psoas, erector spinae muscles, squared lumbar muscles, transverse abdominal muscles, internal and external oblique muscles, and rectus abdominis muscles. The threshold of -190 to -30 HU was used for fat. Visceral fat included all intraperitoneal and retroperitoneal fat, including perirenal. Subcutaneous fat included all fat located in the muscle wall and the skin plane. Image segmentation Code All the used code is open source U-Net was [ 22 ] coded in Python and used the Python libraries numpy [ 23 ], tensorflow [ 24 ], keras [ 25 ], scipy [ 26 ] and imgaug [ 27 ]. The MuViSS ImageJ macros are available at https://github.com/tpecot/MuViSS . Training dataset The training dataset consisted of 320 512 × 512 images while the validation dataset consisted of 12 512 × 512 images. For each image in both datasets, a physician expert manually annotated the backbone, visceral fat, muscle and subcutaneous fat with Annotater [ 28 ]. Training Images were normalized with a 1-99.8 quantile. A root mean square prop was used to estimate the parameters of the deep neural network by minimizing a weighted cross entropy loss to handle class imbalance for 25 epochs. A data augmentation to increase the training dataset by a factor of 100 was processed before normalization with the imgaug python library [ 27 ] and included flipping, pixel dropout, blurring, noise addition and contrast modifications. The trained model in the format that is directly usable with the CSBDeep plugin [ 29 ] in ImageJ is available at https://zenodo.org/record/7990044 . Post-processing 416 images were then segmented with the trained model. An ImageJ macro [ 30 , 31 ] was used to correct segmentations. For both backbone and visceral fat, the largest connected components obtained with the MorpholibJ [ 32 ] plugin followed by hole filling were defined as the segmented areas. For both muscle and subcutaneous fat, the largest connected components obtained with the MorpholibJ [ 32 ] plugin followed by hole filling with a maximum size of 1000 pixels for holes were defined as the segmented areas. Quantification Pixels with an intensity defined between -29 and 150 were set as muscle while pixels with an intensity between - 190 and -30 were set as fat. The area of muscle pixels within the segmented muscle region as well as the area of fat pixels within the segmented subcutaneous and visceral fat regions were measured to compute the visceral fat area. The MuViSS ImageJ macro defined to process this task is available at https://github.com/tpecot/MuViSS . RESULTS In total, the model was trained on 320 images and evaluated on 11 images. Figure 1 shows examples of the final result after analysis by the model. Table 1 summarizes the types of scanners used in this study both for training, validation and evaluation datasets. The accuracy and evaluation metrics of the segmentation are summarized in Table 2 . Note that our model has an accuracy of 97.4%. In order to use our model, video tutorials are available at https://github.com/tpecot/MuViSS . View this table: View inline View popup Download powerpoint Table 1. Brand and type of scanner. View this table: View inline View popup Download powerpoint Table 2. Accuracy and evaluation metrics Download figure Open in new tab Figure 1. Representation of the automatic segmentation by the model. SFA : Sub-Cutaneous Fat Area – SMA : Skeletal Muscle Area – VFA : Visceral Fat Area These tutorials show how to install ImageJ and the required plugins and how to use MuViSS. Note that the intensity for muscle, visceral and subcutaneous fat can be adjusted as input parameters. DISCUSSION Sarcopenia and visceral obesity affect a large number of patients, especially in surgery. Their evaluation is often complex and there is currently no simple tool available for their evaluation. In this study, we demonstrate that a scan-based assessment using Deep Learning was a reliable and reproducible method. This method allows the assessment of the SMA based on the calculation of the area of the L3 level muscles including: the psoas, the erector spinae muscles, the squared lumbar muscles, the transverse abdominal muscles, the internal and external oblique muscles and the rectus abdominis muscles. After adjusting the SMA to the height square, an evaluation of the SMI is made possible. This method also allows to measure, on the same scan section, the VFA represented by the total area of intra-abdominal fat and the SFA represented by the total area of parietal fat. The evaluation of SMI, a reflection of sarcopenia, is widely studied in the literature. Indeed, in oncology, a 10% loss of SMI in patients undergoing radiochemotherapy for esophageal cancer decreased the overall survival of patients [ 33 ]. What’s more, in these patients there is a reduction in progression-free survival and overall survival [ 34 , 35 ]. In the hepatobiliary field, sarcopenia is also a risk factor for 5-year mortality in patients treated for hepatocellular carcinoma. [ 36 ]. In gynecological surgery, sarcopenia decreased overall survival in patients with cervical cancer [ 37 ], endometrial cancer [ 38 , 39 ] and ovarian cancer [ 40 ]. In pancreatic surgery, there is a decrease in recurrence-free survival and overall survival in patients with low SMI after pancreatic resection [ 6 ]. In esophageal surgery, a decrease in survival of sarcopenic patients after esophagectomy has been reported [ 41 ]. In colorectal surgery, sarcopenia associated with an NLR < 3 significantly decreased the overall survival of patients operated on for non-metastatic colorectal cancer [ 42 ]. In liver transplantation, an increased risk of mortality on the liver transplant waiting list has been highlighted in patients with a low SMI. [ 43 ]. The evaluation of obesity is also essential. Indeed, in addition to the specific complications related to obesity such as coronary artery disease, diabetes, etc., obesity, especially visceral obesity, is a risk factor for many complications and a risk factor for mortality. Moreover, although widely used in current practice, BMI is a poor index of fat distribution [ 44 ].The assessment of abdominal obesity, represented by the VFA measurement, is much more relevant. Indeed, in pancreatic surgery, visceral obesity associated with sarcopenia leads to a reduction in recurrence-free survival and overall survival. [ 6 ]. Moreover, abdominal obesity is a risk factor for postoperative infectious complications after pancreatic surgery [ 45 ]. More generally, a meta-analysis summarizes all the postoperative complications associated with abdominal obesity [ 46 ]. In the literature, we find a large number of ways of assessing body composition based on scans : BMI_CT [ 47 , 48 ], Aquarius iNtuition (TeraRecon, San Mateo,CA) [ 49 ], sliceOmatic (TomoVision, Magog, Quebec, Canada) [ 50 , 51 ], Synapse Vincent [ 49 , 50 ], Infinitt PACS [ 51 ]. To date, there has been no significant difference between the different software programs in the assessment of body composition [ 52 ]. The aim of this study is not to demonstrate the superiority of our model, which would require a much more in-depth study. However, the advantage of our method is that it’s free. Indeed, although probably very powerful, tools such as SliceOmatic, Aquarius iNtuition, Synapse Vincent and Osirux require a paid license. BMI_CT, on the other hand, is a semi-quantitative evaluation software, meaning that it requires at least one manual segmentation for each evaluation. The aim was to make available a simple and fast method, free of charge and easy to use. Indeed, our model requires no downloading or special identification, apart from the installation of ImageJ, an open-source software package, or which distributions for Microsoft Windows, Mac OS and Linux are available for download. Our model can therefore be used on any computer. In addition, video tutorials were created to show how to use this tool, available at https://github.com/tpecot/MuViSS . The interest in developing a free, automatic method could, in future, 1) enable systematic assessment of body composition 2) standardize results and thus enable better comparison. To date, no team has developed a deep learning-based access-free assessment model for routine body composition assessment. Moreover, the thresholds, in HU, for the evaluation of intra-abdominal and parietal fat is variable according to the studies. Indeed, some teams use the threshold of -190 to -30 HU while others use thresholds of -150 to -50 HU. In this method, the threshold values retained for fat are adjustable, to allow each person to evaluate the VFA according to the chosen values. CONCLUSION To our knowledge, this is the first study to make available a method for evaluating SMA, VFA and SFA using deep learning in free access, based on ImageJ software. Data Availability All data produced in the present work are contained in the manuscript REFERENCES 1. ↵ Sepúlveda-Loyola W , Osadnik C , Phu S , Morita AA , Duque G , Probst VS . Diagnosis, prevalence, and clinical impact of sarcopenia in COPD: a systematic review and meta-analysis . J Cachexia Sarcopenia Muscle. oct 2020 ; 11 : 1164 – 1176 . doi: 10.1002/jcsm.12600 OpenUrl CrossRef PubMed 2. ↵ Wilkinson TJ , Miksza J , Yates T , Lightfoot CJ , Baker LA , Watson EL , et al. Association of sarcopenia with mortality and end-stage renal disease in those with chronic kidney disease: a UK Biobank study . J Cachexia Sarcopenia Muscle . 2023 ; 12 : 586-598 . doi: 10.1002/jcsm.12705 OpenUrl CrossRef 3. ↵ Tantai X. Effect of sarcopenia on survival in patients with cirrhosis: A meta-analysis . J Hepatol . 2023 ; 76 : 13 . doi: 10.1016/j.jhep.2021.11.006 OpenUrl CrossRef 4. ↵ van Vledder MG , Levolger S , Ayez N , Verhoef C , Tran TCK , IJzermans JNM . Body composition and outcome in patients undergoing resection of colorectal liver metastases . Br J Surg . 2023 ; 8 . doi: 10.1002/bjs.7823 OpenUrl CrossRef PubMed 5. ↵ Abbass T , Tsz Ho YT , Horgan PG , Dolan RD , McMillan DC . The relationship between computed tomography derived skeletal muscle index, psoas muscle index and clinical outcomes in patients with operable colorectal cancer . Clin Nutr ESPEN . 2023 ; 39 : 104 – 113 . doi: 10.1016/j.clnesp.2020.07.010 OpenUrl CrossRef 6. ↵ Okumura S , Kaido T , Hamaguchi Y , Kobayashi A , Shirai H , Yao S , et al. Visceral Adiposity and Sarcopenic Visceral Obesity are Associated with Poor Prognosis After Resection of Pancreatic Cancer . Ann Surg Oncol . 2023 ; 24 : 3732 – 3740 . doi: 10.1245/s10434-017-6077-y OpenUrl CrossRef 7. ↵ Voutilainen A , Brester C , Kolehmainen M , Tuomainen TP . Epidemiological analysis of coronary heart disease and its main risk factors: are their associations multiplicative, additive, or interactive? Ann Med . 2022 ; 54 : 1500 – 1510 . doi: 10.1080/07853890.2022.2078875 OpenUrl CrossRef 8. ↵ Landsberg L , Aronne LJ , Beilin LJ , Burke V , Igel LI , Lloyd-Jones D , et al. Obesity-Related Hypertension: Pathogenesis, Cardiovascular Risk, and Treatment: A Position Paper of The Obesity Society and the American Society of Hypertension . J Clin Hypertens . 2023 ; 15 : 14 – 33 . doi: 10.1111/jch.12049 OpenUrl CrossRef 9. ↵ Flegal KM , Graubard BI , Williamson DF , Gail MH . Cause-Specific Excess Deaths Associated With Underweight, Overweight, and Obesity . JAMA . 2023 ; 298 : 2028 – 2037 . doi: 10.1001/jama.298.17.2028 OpenUrl CrossRef 10. ↵ Kopelman PG . Obesity as a medical problem . Nature . 2023 ; 404 : 635 – 643 . doi: 10.1038/35007508 OpenUrl CrossRef 11. ↵ Iyengar NM , Gucalp A , Dannenberg AJ , Hudis CA . Obesity and Cancer Mechanisms: Tumor Microenvironment and Inflammation . J Clin Oncol . 2023 ; 34 : 4270 – 4276 . doi: 10.1200/JCO.2016.67.4283 OpenUrl CrossRef 12. ↵ Lauby-Secretan B , Scoccianti C , Loomis D , Grosse Y , Bianchini F , Straif K. Body Fatness and Cancer — Viewpoint of the IARC Working Group . N Engl J Med . 2023 ; 375 : 794 – 798 . doi: 10.1056/NEJMsr1606602 OpenUrl CrossRef 13. ↵ Calle EE . Overweight, Obesity, and Mortality from Cancer in a Prospectively Studied Cohort of U.S . Adults. N Engl J Med . 2023 ; 348 : 1625 – 1638 . doi: 10.1056/NEJMoa021423 OpenUrl CrossRef 14. ↵ Powell-Wiley TM , Poirier P , Burke LE , Després JP , Gordon-Larsen P , Lavie CJ , et al. Obesity and Cardiovascular Disease: A Scientific Statement From the American Heart Association . Circulation . 2023 ; 143 . doi: 10.1161/CIR.0000000000000973 OpenUrl CrossRef PubMed 15. ↵ Mitsiopoulos N , Baumgartner RN , Heymsfield SB , Lyons W , Gallagher D , Ross R. Cadaver validation of skeletal muscle measurement by magnetic resonance imaging and computerized tomography . J Appl Physiol . 2023 ; 85 : 115 – 122 . doi: 10.1152/jappl.1998.85.1.115 OpenUrl CrossRef 16. Heymsfield SB , Adamek M , Gonzalez MC , Jia G , Thomas DM . Assessing skeletal muscle mass: historical overview and state of the art . J Cachexia Sarcopenia Muscle . 2023 ; 5 : 9 – 18 . doi: 10.1007/s13539-014-0130-5 OpenUrl CrossRef 17. Mourtzakis M , Prado CMM , Lieffers JR , Reiman T , McCargar LJ , Baracos VE . A practical and precise approach to quantification of body composition in cancer patients using computed tomography images acquired during routine care . Appl Physiol Nutr Metab . 2023 ; 33 : 997 – 1006 . doi: 10.1139/H08-075 OpenUrl CrossRef 18. ↵ Shen W , Punyanitya M , Wang Z , Gallagher D , St.-Onge MP , Albu J , et al. Total body skeletal muscle and adipose tissue volumes: estimation from a single abdominal cross-sectional image . J Appl Physiol . 2023 ; 97 : 2333 – 2338 . doi: 10.1152/japplphysiol.00744.2004 OpenUrl CrossRef 19. ↵ Derstine BA , Holcombe SA , Ross BE , Wang NC , Su GL , Wang SC . Skeletal muscle cutoff values for sarcopenia diagnosis using T10 to L5 measurements in a healthy US population . Sci Rep . 2023 ; 8 : 11369 . doi: 10.1038/s41598-018-29825-5 OpenUrl CrossRef 20. ↵ van der Werf A , Langius JAE , de van der Schueren MAE , Nurmohamed SA , van der Pant KAMI , Blauwhoff-Buskermolen S , et al. Percentiles for skeletal muscle index, area and radiation attenuation based on computed tomography imaging in a healthy Caucasian population . Eur J Clin Nutr . 2023 ; 72 : 288 – 296 . doi: 10.1038/s41430-017-0034-5 OpenUrl CrossRef 21. ↵ Examination Committee of Criteria for ‘Obesity Disease’ in Japan . New Criteria for ‘Obesity Disease’ in Japan . Circ J . 2023 ; 6 . doi: 10.1253/circj.66.987 OpenUrl CrossRef PubMed Web of Science 22. ↵ Navab N , Hornegger J , Wells WM , Frangi AF Ronneberger O , Fischer P , Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation . In: Navab N , Hornegger J , Wells WM , Frangi AF , éditeurs. Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015 . Springer International . 2015 : 234 – 241 . OpenUrl 23. ↵ Van Der Walt S , Colbert SC , Varoquaux G. The NumPy array: a structure for efficient numerical computation . Comput Sci Eng . 2011 ; 13 : 22 – 30 . OpenUrl CrossRef PubMed 24. ↵ Abadi M , Agarwal A , Barham P , Brevdo E , Chen Z , Citro C , et al. TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems . arXiv ; 2016 . 25. ↵ Chollet F. Keras , 2015 . Available at https://github.com/fchollet/keras . 26. ↵ Virtanen P , Gommers R , Oliphant TE , Haberland M , Reddy T , Cournapeau D , et al. SciPy 1.0-Fundamental Algorithms for Scientific Computing in Python . Nat Methods . 2023 ; 17 : 261 – 272 . doi: 10.1038/s41592-019-0686-2 OpenUrl CrossRef PubMed 27. ↵ Jung AB , Wada K , Crall J , Tanaka S , Graving J , Reinders C , et al. imgaug , 2020 . Available at https://github.com/aleju/imgaug . 28. ↵ Pécot T , Cuitiño MC , Johnson RH , Timmers C , Leone G. Deep learning tools and modeling to estimate the temporal expression of cell cycle proteins from 2D still images . PLOS Comput Biol . 2023 ; 18 : e1009949 . doi: 10.1371/journal.pcbi.1009949 OpenUrl CrossRef 29. ↵ Weigert M , Schmidt U , Boothe T , Müller A , Dibrov A , Jain A , et al. Content-aware image restoration: pushing the limits of fluorescence microscopy . Nature methods . 2023 ; 15 : 1090 – 1097 . doi: 10.1038/s41592-018-0216-7 OpenUrl CrossRef 30. ↵ Yoon HG , Oh D , Ahn YC , Noh JM , Pyo H , Cho WK , et al. Prognostic Impact of Sarcopenia and Skeletal Muscle Loss During Neoadjuvant Chemoradiotherapy in Esophageal Cancer . Cancers . 2023 ; 12 : 925 . doi: 10.3390/cancers12040925 OpenUrl CrossRef 31. ↵ Takada K , Yoneshima Y , Tanaka K , Okamoto I , Shimokawa M , Wakasu S , et al. Clinical impact of skeletal muscle area in patients with non-small cell lung cancer treated with anti-PD-1 inhibitors . J Cancer Res Clin Oncol . 2023 ; 146 : 1217 – 1225 . doi: 10.1007/s00432-020-03146-5 OpenUrl CrossRef 32. ↵ Cho WK , Yu JI , Park HC , Lim DH , Kim TH , Chie EK . Impact of sarcopenia on survival of pancreatic cancer patients treated with concurrent chemoradiotherapy . Tumori J . 2023 ; 107 : 247 – 253 . doi: 10.1177/0300891620937795 OpenUrl CrossRef 33. ↵ Fujiwara N. Sarcopenia , Intramuscular Fat Deposition, and Visceral Adiposity Independently Predict the Outcomes of Hepatocellular Carcinoma . J Hepatol . 2023 ; 63 : 131 – 140 . doi: 10.1016/j.jhep.2015.02.031 OpenUrl CrossRef 34. ↵ Lee J , Chang CL , Lin JB , Wu MH , Sun FJ , Jan YT , et al. Skeletal Muscle Loss Is an Imaging Biomarker of Outcome after Definitive Chemoradiotherapy for Locally Advanced Cervical Cancer . Clin Cancer Res . 2023 ; 24 : 5028 – 5036 . doi: 10.1158/1078-0432.CCR-18-0788 OpenUrl CrossRef 35. ↵ De Paula NS , Rodrigues CS , Chaves GV . Comparison of the prognostic value of different skeletal muscle radiodensity parameters in endometrial cancer . Eur J Clin Nutr . 2023 ; 73 : 524 – 530 . doi: 10.1038/s41430-018-0163-5 OpenUrl CrossRef 36. ↵ Rodrigues CS , Chaves GV . Skeletal Muscle Quality Beyond Average Muscle Attenuation: A Proposal of Skeletal Muscle Phenotypes to Predict Short-Term Survival in Patients With Endometrial Cancer . J Natl Compr Canc Netw . 2023 ; 16 : 153 – 160 . doi: 10.6004/jnccn.2017.7028 OpenUrl CrossRef 37. ↵ Rutten IJG , Van Dijk DPJ , Kruitwagen RFPM , Beets-Tan RGH , Olde Damink SWM , Van Gorp T. Loss of skeletal muscle during neoadjuvant chemotherapy is related to decreased survival in ovarian cancer patients: Loss of skeletal muscle in ovarian cancer . J Cachexia Sarcopenia Muscle . 2023 ; 7 : 458 – 466 . doi: 10.1002/jcsm.12107 OpenUrl CrossRef 38. ↵ Srpcic M , Jordan T , Popuri K , Sok M. Sarcopenia and myosteatosis at presentation adversely affect survival after esophagectomy for esophageal cancer . Radiol Oncol . 2023 ; 54 : 237 – 246 . doi: 10.2478/raon-2020-0016 OpenUrl CrossRef 39. ↵ Cespedes Feliciano EM , Kroenke CH , Meyerhardt JA , Prado CM , Bradshaw PT , Kwan ML , et al. Association of Systemic Inflammation and Sarcopenia With Survival in Nonmetastatic Colorectal Cancer: Results From the C SCANS Study . JAMA Oncol . 2023 ; 3 : e172319 . doi: 10.1001/jamaoncol.2017.2319 OpenUrl CrossRef 40. ↵ Bot D , Droop A , Lucassen CJ , Van Veen ME , Van Vugt JLA , Shahbazi Feshtali S , et al. Both muscle quantity and quality are predictors of waiting list mortality in patients with end-stage liver disease . Clin Nutr ESPEN . 2023 ; 42 : 272 – 279 . doi: 10.1016/j.clnesp.2021.01.022 OpenUrl CrossRef 41. ↵ Rickles AS , Iannuzzi JC , Mironov O , Deeb AP , Sharma A , Fleming FJ , et al. Visceral Obesity and Colorectal Cancer: Are We Missing the Boat with BMI? J Gastrointest Surg . 2013 ; 11 . doi: 10.1007/s11605-012-2045-9 OpenUrl CrossRef PubMed 42. ↵ Van Dijk DPJ , Bakens MJAM , Coolsen MME , Rensen SS , Van Dam RM , Bours MJL , et al. Low skeletal muscle radiation attenuation and visceral adiposity are associated with overall survival and surgical site infections in patients with pancreatic cancer: Muscle radiation attenuation in pancreatic cancer . J Cachexia Sarcopenia Muscle . 2023 ; 8 : 317 – 326 . doi: 10.1002/jcsm.12155 OpenUrl CrossRef 43. ↵ Saravana-Bawan B , Goplen M , Alghamdi M , Khadaroo RG . The Relationship Between Visceral Obesity and Post-operative Complications: A Meta-Analysis . J Surg Res . 2023 ; 267 : 71 – 81 . doi: 10.1016/j.jss.2021.04.034 OpenUrl CrossRef 44. ↵ Kang SH , Jeong WK , Baik SK , Cha SH , Kim MY . Impact of sarcopenia on prognostic value of cirrhosis: going beyond the hepatic venous pressure gradient and MELD score: Impact of sarcopenia on prognostic value of cirrhosis . J Cachexia Sarcopenia Muscle . 2023 ; 9 : 860 – 870 . doi: 10.1002/jcsm.12333 OpenUrl CrossRef 45. ↵ Hou L , Deng Y , Fan X , Zhao T , Cui B , Lin L , et al. A Sex-Stratified Prognostic Nomogram Incorporating Body Compositions for Long-Term Mortality in Cirrhosis . J Parenter Enter Nutr . 2023 ; 45 : 403 – 413 . doi: 10.1002/jpen.1841 OpenUrl CrossRef 46. ↵ Hamaguchi Y , Kaido T , Okumura S , Kobayashi A , Shirai H , Yao S , et al. Including body composition in MELD scores improves mortality prediction among patients awaiting liver transplantation . Clin Nutr . 2023 ; 39 : 1885 – 1892 . doi: 10.1016/j.clnu.2019.08.012 OpenUrl CrossRef 47. ↵ Hanai T , Shiraki M , Nishimura K , Ohnishi S , Imai K , Suetsugu A , et al. Sarcopenia impairs prognosis of patients with liver cirrhosis . Nutrition . 2023 ; 31 : 193 – 199 . doi: 10.1016/j.nut.2014.07.005 OpenUrl CrossRef 48. ↵ Lattanzi B , Nardelli S , Pigliacelli A , Di Cola S , Farcomeni A , D’Ambrosio D , et al. The additive value of sarcopenia, myosteatosis and hepatic encephalopathy in the predictivity of model for end-stage liver disease . Dig Liver Dis . 2023 ; 51 : 1508 – 1512 . doi: 10.1016/j.dld.2019.09.004 OpenUrl CrossRef 49. ↵ Sakurai K , Kubo N , Tamura T , Toyokawa T , Amano R , Tanaka H , et al. Adverse Effects of Low Preoperative Skeletal Muscle Mass in Patients Undergoing Gastrectomy for Gastric Cancer . Ann Surg Oncol . 2023 ; 24 : 2712 – 2719 . doi: 10.1245/s10434-017-5875-6 OpenUrl CrossRef 50. ↵ Tanaka K , Yamada S , Sonohara F , Takami H , Hayashi M , Kanda M , et al. Pancreatic Fat and Body Composition Measurements by Computed Tomography are Associated with Pancreatic Fistula After Pancreatectomy . Ann Surg Oncol . 2023 ; 28 : 530 – 538 . doi: 10.1245/s10434-020-08581-9 OpenUrl CrossRef 51. ↵ Zhuang CL , Huang DD , Pang WY , Zhou CJ , Wang SL , Lou N , et al. Sarcopenia is an Independent Predictor of Severe Postoperative Complications and Long-Term Survival After Radical Gastrectomy for Gastric Cancer: Analysis from a Large-Scale Cohort . Medicine ( Baltimore ). 2023 ; 95 : e3164 . doi: 10.1097/MD.0000000000003164 OpenUrl CrossRef 52. ↵ van Vugt JLA , Levolger S , Gharbharan A , Koek M , Niessen WJ , Burger JWA , et al. A comparative study of software programmes for cross-sectional skeletal muscle and adipose tissue measurements on abdominal computed tomography scans of rectal cancer patients: Software programmes for body composition measurements on CT . J Cachexia Sarcopenia Muscle . 2017 ; 8 : 285 – 297 . doi: 10.1002/jcsm.12158 OpenUrl CrossRef View the discussion thread. Back to top Previous Next Posted March 13, 2024. Download PDF Data/Code Email Thank you for your interest in spreading the word about medRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following MuViSS : Muscle, Visceral and Subcutaneous Segmentation by an automatic evaluation method using Deep Learning Message Subject (Your Name) has forwarded a page to you from medRxiv Message Body (Your Name) thought you would like to see this page from the medRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share MuViSS : Muscle, Visceral and Subcutaneous Segmentation by an automatic evaluation method using Deep Learning Edouard Wasielewski , Karim Boudjema , Laurent Sulpice , Thierry Pecot medRxiv 2024.03.11.24304074; doi: https://doi.org/10.1101/2024.03.11.24304074 Share This Article: Copy Citation Tools MuViSS : Muscle, Visceral and Subcutaneous Segmentation by an automatic evaluation method using Deep Learning Edouard Wasielewski , Karim Boudjema , Laurent Sulpice , Thierry Pecot medRxiv 2024.03.11.24304074; doi: https://doi.org/10.1101/2024.03.11.24304074 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Surgery Subject Areas All Articles Addiction Medicine (573) Allergy and Immunology (865) Anesthesia (302) Cardiovascular Medicine (4453) Dentistry and Oral Medicine (444) Dermatology (383) Emergency Medicine (609) Endocrinology (including Diabetes Mellitus and Metabolic Disease) (1515) Epidemiology (15242) Forensic Medicine (30) Gastroenterology (1131) Genetic and Genomic Medicine (6615) Geriatric Medicine (669) Health Economics (1001) Health Informatics (4552) Health Policy (1372) Health Systems and Quality Improvement (1614) Hematology (543) HIV/AIDS (1270) Infectious Diseases (except HIV/AIDS) (15929) Intensive Care and Critical Care Medicine (1106) Medical Education (624) Medical Ethics (147) Nephrology (670) Neurology (6625) Nursing (346) Nutrition (999) Obstetrics and Gynecology (1148) Occupational and Environmental Health (957) Oncology (3344) Ophthalmology (979) Orthopedics (369) Otolaryngology (421) Pain Medicine (436) Palliative Medicine (130) Pathology (665) Pediatrics (1696) Pharmacology and Therapeutics (693) Primary Care Research (714) Psychiatry and Clinical Psychology (5461) Public and Global Health (9252) Radiology and Imaging (2207) Rehabilitation Medicine and Physical Therapy (1371) Respiratory Medicine (1197) Rheumatology (597) Sexual and Reproductive Health (715) Sports Medicine (530) Surgery (714) Toxicology (99) Transplantation (289) Urology (265) (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'a02de4a8d82b8e2e',t:'MTc3OTk3Nzk1NA=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();