Automated segmentation of muscle and adipose tissue on CT images for human body composition analysis

Author(s):  
Howard Chung ◽  
Dana Cobzas ◽  
Laura Birdsell ◽  
Jessica Lieffers ◽  
Vickie Baracos
2002 ◽  
Vol 26 (1) ◽  
pp. 21-29 ◽  
Author(s):  
Paul R. Buzzell ◽  
Valerie M. Chamberlain ◽  
Stephen J. Pintauro

This study examined the effectiveness of a series of Web-based, multimedia tutorials on methods of human body composition analysis. Tutorials were developed around four body composition topics: hydrodensitometry (underwater weighing), dual-energy X-ray absorptiometry, bioelectrical impedance analysis, and total body electrical conductivity. Thirty-two students enrolled in the course were randomly assigned to learn the material through either the Web-based tutorials only (“Computer”), a traditional lecture format (“Lecture”), or lectures supplemented with Web-based tutorials (“Both”). All students were administered a validated pretest before randomization and an identical posttest at the completion of the course. The reliability of the test was 0.84. The mean score changes from pretest to posttest were not significantly different among the groups (65.4 ± 17.31, 78.82 ± 21.50, and 76 ± 21.22 for the Computer, Both, and Lecture groups, respectively). Additionally, a Likert-type assessment found equally positive attitudes toward all three formats. The results indicate that Web-based tutorials are as effective as the traditional lecture format for teaching these topics.


2020 ◽  
Vol 47 (11) ◽  
pp. 5723-5730
Author(s):  
Yabo Fu ◽  
Joseph E. Ippolito ◽  
Daniel R. Ludwig ◽  
Rehan Nizamuddin ◽  
Harold H. Li ◽  
...  

2020 ◽  
Vol 8 (2) ◽  
pp. e000821
Author(s):  
Arissa C Young ◽  
Henry T Quach ◽  
Haocan Song ◽  
Elizabeth J Davis ◽  
Javid J Moslehi ◽  
...  

BackgroundImmune checkpoint inhibitors (ICIs) have transformed treatment for melanoma, but identifying reliable biomarkers of response and effective modifiable lifestyle factors has been challenging. Obesity has been correlated with improved responses to ICI, although the association of body composition measures (muscle, fat, etc) with outcomes remains unknown.MethodsWe performed body composition analysis using Slice-o-matic software on pretreatment CT scans to quantify skeletal muscle index (SMI=skeletal muscle area/height2), skeletal muscle density (SMD), skeletal muscle gauge (SMG=SMI × SMD), and total adipose tissue index (TATI=subcutaneous adipose tissue area + visceral adipose tissue area/height2) of each patient at the third lumbar vertebrae. We then correlated these measures to response, progression-free survival (PFS), overall survival (OS), and toxicity.ResultsAmong 287 patients treated with ICI, body mass index was not associated with clinical benefit or toxicity. In univariable analyses, patients with sarcopenic obesity had inferior PFS (HR 1.4, p=0.04). On multivariable analyses, high TATI was associated with inferior PFS (HR 1.7, p=0.04), which was particularly strong in women (HR 2.1, p=0.03). Patients with intermediate TATI and high SMG had the best outcomes, whereas those with low SMG/high TATI had inferior PFS and OS (p=0.02 for both PFS and OS).ConclusionsBody composition analysis identified several features that correlated with improved clinical outcomes, although the associations were modest. As with other studies, we identified sex-specific associations that warrant further study.


2021 ◽  
Vol 5 (1) ◽  
Author(s):  
Pablo Borrelli ◽  
Reza Kaboteh ◽  
Olof Enqvist ◽  
Johannes Ulén ◽  
Elin Trägårdh ◽  
...  

Abstract Background Body composition is associated with survival outcome in oncological patients, but it is not routinely calculated. Manual segmentation of subcutaneous adipose tissue (SAT) and muscle is time-consuming and therefore limited to a single CT slice. Our goal was to develop an artificial-intelligence (AI)-based method for automated quantification of three-dimensional SAT and muscle volumes from CT images. Methods Ethical approvals from Gothenburg and Lund Universities were obtained. Convolutional neural networks were trained to segment SAT and muscle using manual segmentations on CT images from a training group of 50 patients. The method was applied to a separate test group of 74 cancer patients, who had two CT studies each with a median interval between the studies of 3 days. Manual segmentations in a single CT slice were used for comparison. The accuracy was measured as overlap between the automated and manual segmentations. Results The accuracy of the AI method was 0.96 for SAT and 0.94 for muscle. The average differences in volumes were significantly lower than the corresponding differences in areas in a single CT slice: 1.8% versus 5.0% (p < 0.001) for SAT and 1.9% versus 3.9% (p < 0.001) for muscle. The 95% confidence intervals for predicted volumes in an individual subject from the corresponding single CT slice areas were in the order of ± 20%. Conclusions The AI-based tool for quantification of SAT and muscle volumes showed high accuracy and reproducibility and provided a body composition analysis that is more relevant than manual analysis of a single CT slice.


2019 ◽  
Vol 75 ◽  
pp. 47-55 ◽  
Author(s):  
Setareh Dabiri ◽  
Karteek Popuri ◽  
Elizabeth M. Cespedes Feliciano ◽  
Bette J. Caan ◽  
Vickie E. Baracos ◽  
...  

2020 ◽  
Vol 55 (6) ◽  
pp. 357-366
Author(s):  
Sebastian Nowak ◽  
Anton Faron ◽  
Julian A. Luetkens ◽  
Helena L. Geißler ◽  
Michael Praktiknjo ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document