Michael T Paris1, Puneeta Tandon2, Daren K Heyland3, Helena Furberg4, Tahira Premji1, Gavin Low5, Marina Mourtzakis6. 1. Department of Kinesiology, University of Waterloo, Waterloo, ON, Canada. 2. Department of Gastroenterology, University of Alberta, Edmonton, AB, Canada. 3. Department of Critical Care, Kingston General Hospital, Kingston, ON, Canada; Clinical Evaluation Research Unit, Queens University, Kingston, ON, Canada. 4. Department of Epidemiology and Biostatistics, Memorial Sloan Kettering Cancer Center, New York, NY, USA. 5. Department of Radiology, University of Alberta, Edmonton, AB, Canada. 6. Department of Kinesiology, University of Waterloo, Waterloo, ON, Canada. Electronic address: mmourtzakis@uwaterloo.ca.
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.
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 carcinomapatients, 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.
Authors: Alexander D Weston; Panagiotis Korfiatis; Timothy L Kline; Kenneth A Philbrick; Petro Kostandy; Tomas Sakinis; Motokazu Sugimoto; Naoki Takahashi; Bradley J Erickson Journal: Radiology Date: 2018-12-11 Impact factor: 11.105
Authors: Katie M Di Sebastiano; Lin Yang; Kevin Zbuk; Raimond K Wong; Tom Chow; David Koff; Gerald R Moran; Marina Mourtzakis Journal: Br J Nutr Date: 2012-07-04 Impact factor: 3.718
Authors: Lisa Martin; Jessica Hopkins; Georgios Malietzis; J T Jenkins; Michael B Sawyer; Ron Brisebois; Anthony MacLean; Gregg Nelson; Leah Gramlich; Vickie E Baracos Journal: Ann Surg Oncol Date: 2018-07-13 Impact factor: 5.344
Authors: Maria Cristina Gonzalez; Maria Isabel T D Correia; Steven B Heymsfield Journal: Curr Opin Clin Nutr Metab Care Date: 2017-09 Impact factor: 4.294
Authors: Ali N Gunesch; Thomas L Sutton; Stephanie M Krasnow; Christopher R Deig; Brett C Sheppard; Daniel L Marks; Aaron J Grossberg Journal: Am J Surg Date: 2022-03-26 Impact factor: 3.125
Authors: Kaushalya C Amarasinghe; Jamie Lopes; Julian Beraldo; Nicole Kiss; Nicholas Bucknell; Sarah Everitt; Price Jackson; Cassandra Litchfield; Linda Denehy; Benjamin J Blyth; Shankar Siva; Michael MacManus; David Ball; Jason Li; Nicholas Hardcastle Journal: Front Oncol Date: 2021-05-07 Impact factor: 6.244
Authors: Jennifer C Lai; Puneeta Tandon; William Bernal; Elliot B Tapper; Udeme Ekong; Srinivasan Dasarathy; Elizabeth J Carey Journal: Hepatology Date: 2021-09 Impact factor: 17.298
Authors: Leanne L G C Ackermans; Leroy Volmer; Leonard Wee; Ralph Brecheisen; Patricia Sánchez-González; Alexander P Seiffert; Enrique J Gómez; Andre Dekker; Jan A Ten Bosch; Steven M W Olde Damink; Taco J Blokhuis Journal: Sensors (Basel) Date: 2021-03-16 Impact factor: 3.576