Building Large-Scale Quantitative Imaging Databases with Multi-Scale Deep Reinforcement Learning: Initial Experience with Whole-Body Organ Volumetric Analyses

Journal of Digital Imaging - Tập 34 - Trang 124-133 - 2021
David J. Winkel1, Hanns-Christian Breit1, Thomas J. Weikert1, Bram Stieltjes1
1Department of Radiology, University Hospital of Basel, Basel, Switzerland

Tóm tắt

To explore the feasibility of a fully automated workflow for whole-body volumetric analyses based on deep reinforcement learning (DRL) and to investigate the influence of contrast-phase (CP) and slice thickness (ST) on the calculated organ volume. This retrospective study included 431 multiphasic CT datasets—including three CP and two ST reconstructions for abdominal organs—totaling 10,508 organ volumes (10,344 abdominal organ volumes: liver, spleen, and kidneys, 164 lung volumes). Whole-body organ volumes were determined using multi-scale DRL for 3D anatomical landmark detection and 3D organ segmentation. Total processing time for all volumes and mean calculation time per case were recorded. Repeated measures analyses of variance (ANOVA) were conducted to test for robustness considering CP and ST. The algorithm calculated organ volumes for the liver, spleen, and right and left kidney (mean volumes in milliliter (interquartile range), portal venous CP, 5 mm ST: 1868.6 (1426.9, 2157.8), 350.19 (45.46, 395.26), 186.30 (147.05, 214.99) and 181.91 (143.22, 210.35), respectively), and for the right and left lung (2363.1 (1746.3, 2851.3) and 1950.9 (1335.2, 2414.2)). We found no statistically significant effects of the variable contrast phase or the variable slice thickness on the organ volumes. Mean computational time per case was 10 seconds. The evaluated approach, using state-of-the art DRL, enables a fast processing of substantial amounts irrespective of CP and ST, allowing building up organ-specific volumetric databases. The thus derived volumes may serve as reference for quantitative imaging follow-up.

Tài liệu tham khảo

Fitzpatrick JA, Kim JU, Cobbold JFL, et al. Changes in Liver Volume in Patients with Chronic Hepatitis C Undergoing Antiviral Therapy. J Clin Exp Hepatol. 2016;6(1):15–20.

Hayashi T, Saitoh S, Fukuzawa K, et al. Noninvasive Assessment of Advanced Fibrosis Based on Hepatic Volume in Patients with Nonalcoholic Fatty Liver Disease. Gut Liver. 2017/06/27. Editorial Office of Gut and Liver; 2017;11(5):674–683 https://www.ncbi.nlm.nih.gov/pubmed/28651300.

Yamagishi Y, Saito H, Ebinuma H, et al. A new prognostic formula for adult acute liver failure using computer tomography-derived hepatic volumetric analysis. J Gastroenterol. 2009;44(6):615–623.

Veroux M, Gozzo C, Corona D, et al. Change in kidney volume after kidney transplantation in patients with autosomal polycystic kidney disease. PLoS One. Public Library of Science; 2018;13(12):e0209332–e0209332 https://www.ncbi.nlm.nih.gov/pubmed/30589879.

Linguraru MG, Sandberg JK, Jones EC, Summers RM. Assessing splenomegaly: automated volumetric analysis of the spleen. Acad Radiol. 2013/03/25. 2013;20(6):675–684 https://www.ncbi.nlm.nih.gov/pubmed/23535191.

Tenda ED, Ridge CA, Shen M, Yang GZ, Shah PL. Role of Quantitative Computed Tomographic Scan Analysis in Lung Volume Reduction for Emphysema. Respiration. 2019;98(1):86–94.

Frericks BB, Caldarone FC, Nashan B, et al. 3D CT modeling of hepatic vessel architecture and volume calculation in living donated liver transplantation. Eur Radiol. 2004;14(2):326–333.

Gotra A, Sivakumaran L, Chartrand G, et al. Liver segmentation: indications, techniques and future directions. Insights Imaging. 2017;8(4):377–392.

Winkel DJ, Weikert TJ, Breit H-C, et al. Validation of a fully automated liver segmentation algorithm using multi-scale deep reinforcement learning and comparison versus manual segmentation. Eur J Radiol. Elsevier; 2020;126 https://doi.org/10.1016/j.ejrad.2020.108918.

Ghesu FC, Georgescu B, Zheng Y, et al. Multi-Scale Deep Reinforcement Learning for Real-Time 3D-Landmark Detection in CT Scans. IEEE Trans Pattern Anal Mach Intell. 2019;41(1):176–189.

Ghesu FC, Georgescu B, Zheng Y, et al. Multi-Scale Deep Reinforcement Learning for Real-Time 3D-Landmark Detection in CT Scans. IEEE Trans Pattern Anal Mach Intell. 2017;14(2):176–189.

Sutton RS, Barto AG. Introduction to reinforcement learning. Cambride MIT Press. 1998;2(4).

Ghesu FC, Georgescu B, Grbic S, Maier A, Hornegger J, Comaniciu D. Robust multi-scale anatomical landmark detection in incomplete 3D-CT data. Int Conf Med Image Comput Comput Interv. 2017. p. 194–202.

Yang D, Xu D, Zhou SK, et al. Automatic liver segmentation using an adversarial image-to-image network. Descoteaux M, Maier-Hein L, Franz A, Jannin P, Collins DL, Duchesne S, Ed Med Image Comput Comput Assist Interv − MICCAI 2017 Cham Springer Int Publ Cham Springer Int Publ. 2017. p. 507–515.

Kucybała I, Ciuk S, Tęczar J. Spleen enlargement assessment using computed tomography: which coefficient correlates the strongest with the real volume of the spleen? Abdom Radiol (New York). Springer US; 2018;43(9):2455–2461 https://pubmed.ncbi.nlm.nih.gov/29460042.

Monsky WL, Garza AS, Kim I, et al. Treatment planning and volumetric response assessment for Yttrium-90 radioembolization: semiautomated determination of liver volume and volume of tumor necrosis in patients with hepatic malignancy. Cardiovasc Intervent Radiol. 2010/08/04. Springer-Verlag; 2011;34(2):306–318 https://www.ncbi.nlm.nih.gov/pubmed/20683722.

Kawel-Boehm N, Maceira A, Valsangiacomo-Buechel ER, et al. Normal values for cardiovascular magnetic resonance in adults and children. J Cardiovasc Magn Reson. 2015;17(1):29 https://doi.org/10.1186/s12968-015-0111-7.

Neill DB. Using artificial intelligence to improve hospital inpatient care. IEEE Intell Syst. 2013.

Mnih V, Kavukcuoglu K, Silver D, et al. Human-level control through deep reinforcement learning. Nature. 2015;518(7549):529–533.

Prassopoulos P, Daskalogiannaki M, Raissaki M, Hatjidakis A, Gourtsoyiannis N. Determination of normal splenic volume on computed tomography in relation to age, gender and body habitus. Eur Radiol. 1997;7(2):246–248 https://doi.org/10.1007/s003300050145.

Harris A, Kamishima T, Hao HY, et al. Splenic volume measurements on computed tomography utilizing automatically contouring software and its relationship with age, gender, and anthropometric parameters. Eur J Radiol. 2010;75(1):e97–e101 http://www.sciencedirect.com/science/article/pii/S0720048X0900504X.

Cheong B, Muthupillai R, Rubin MF, Flamm SD. Normal values for renal length and volume as measured by magnetic resonance imaging. Clin J Am Soc Nephrol. 2007;2(1):38–45.

Haas M, Hamm B, Niehues SM. Automated lung volumetry from routine thoracic CT scans: How reliable is the result? Acad Radiol. Elsevier Ltd; 2014;21(5):633–638 https://doi.org/10.1016/j.acra.2014.01.002.

Graffy PM, Sandfort V, Summers RM, Pickhardt PJ. Automated liver fat quantification at nonenhanced abdominal CT for population-based steatosis assessment. Radiology. Radiological Society of North America; 2019;293(2):334–342 https://doi.org/10.1148/radiol.2019190512.

Thông tin
Thông tin xuất bản

Journal of Digital Imaging

Tập 34

124-133

Thông tin tác giả