Automated body composition analysis of clinically acquired computed tomography scans using neural networks Journal Article
| Authors: | Paris, M. T.; Tandon, P.; Heyland, D. K.; Furberg, H.; Premji, T.; Low, G.; Mourtzakis, M. |
| Article Title: | Automated body composition analysis of clinically acquired computed tomography scans using neural networks |
| Abstract: | Background & aims: The quantity and quality of skeletal muscle and adipose tissue is an important prognostic factor for clinical outcomes across several illnesses. Clinically acquired computed tomography (CT) scans are commonly used for quantification of body composition, but manual analysis is laborious and costly. The primary aim of this study was to develop an automated body composition analysis framework using CT scans. Methods: CT scans of the 3rd lumbar vertebrae from critically ill, liver cirrhosis, pancreatic cancer, and clear cell renal cell carcinoma patients, as well as renal and liver donors, were manually analyzed for body composition. Ninety percent of scans were used for developing and validating a neural network for the automated segmentation of skeletal muscle and adipose tissues. Network accuracy was evaluated with the remaining 10 percent of scans using the Dice similarity coefficient (DSC), which quantifies the overlap (0 = no overlap, 1 = perfect overlap) between human and automated segmentations. Results: Of the 893 patients, 44% were female, with a mean (±SD) age and body mass index of 52.7 (±15.8) years old and 28.0 (±6.1) kg/m2, respectively. In the testing cohort (n = 89), DSC scores indicated excellent agreement between human and network-predicted segmentations for skeletal muscle (0.983 ± 0.013), and intermuscular (0.900 ± 0.034), visceral (0.979 ± 0.019), and subcutaneous (0.986 ± 0.016) adipose tissue. Network segmentation took ~350 milliseconds/scan using modern computing hardware. Conclusions: Our network displayed excellent ability to analyze diverse body composition phenotypes and clinical cohorts, which will create feasible opportunities to advance our capacity to predict health outcomes in clinical populations. © 2020 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism |
| Keywords: | adult; middle aged; major clinical study; liver cirrhosis; pancreas cancer; computer assisted tomography; cohort analysis; renal cell carcinoma; body mass; body composition; computed tomography; adipose tissue; skeletal muscle; critically ill patient; image segmentation; neural network; clinical outcome; intra-abdominal fat; subcutaneous fat; lumbar vertebra; muscle tissue; autoanalysis; kidney donor; measurement accuracy; human; male; female; article; sarcopenia; convolutional neural network; automated body composition analysis; liver donor |
| Journal Title: | Clinical Nutrition |
| Volume: | 39 |
| Issue: | 10 |
| ISSN: | 0261-5614 |
| Publisher: | Elsevier B.V. |
| Date Published: | 2020-10-01 |
| Start Page: | 3049 |
| End Page: | 3055 |
| Language: | English |
| DOI: | 10.1016/j.clnu.2020.01.008 |
| PUBMED: | 32007318 |
| PROVIDER: | scopus |
| PMCID: | PMC7374050 |
| DOI/URL: | |
| Notes: | Article -- Export Date: 2 November 2020 -- Source: Scopus |
Altmetric
Citation Impact
BMJ Impact Analytics
MSK Authors
Related MSK Work