Top

07-12-2024 | Biomarkers | Research

Comparing fully automated AI body composition biomarkers at differing virtual monoenergetic levels using dual-energy CT

Authors: Giuseppe V. Toia, John W. Garret, Sean D. Rose, Timothy P. Szczykutowicz, Perry J. Pickhardt

Abstract

Purpose

To investigate the behavior of artificial intelligence (AI) CT-based body composition biomarkers at different virtual monoenergetic imaging (VMI) levels using dual-energy CT (DECT).

Methods

This retrospective study included 88 contrast-enhanced abdominopelvic CTs acquired with rapid-kVp switching DECT. Images were reconstructed into five VMI levels (40, 55, 70, 85, 100 keV). Fully automated algorithms for quantifying CT number (HU) in abdominal fat (subcutaneous and visceral), skeletal muscle, bone, calcium (abdominal Agatston score), and organ size (area or volume) were applied. Biomarker median difference relative to 70 keV and interquartile range were reported by energy level to characterize variation. Linear regression was performed to calibrate non-70 keV data and to estimate their equivalent 70 keV biomarker attenuation values.

Results

Relative to 70 keV, absolute median differences in attenuation-based biomarkers (excluding Agatston score) ranged 39–358, 12–102, 5–48, 9–75 HU for 40, 55, 85, 100 keV, respectively. For area-based biomarkers, differences ranged 6–15, 3–4, 2–7, 0–5 cm² for 40, 55, 85, 100 keV. For volume-based biomarkers, differences ranged 12–34, 8–68, 12–52, 1–57 cm³ for 40, 55, 85, 100 keV. Agatston score behavior was more spurious with median differences ranging 70–204 HU. In general, VMI < 70 keV showed more variation in median biomarker measurement than VMI > 70 keV.

Conclusion

This study characterized the behavior of a fully automated AI CT biomarker toolkit across varying VMI levels obtained with DECT. The data showed relatively little biomarker value change when measured at or greater than 70 keV. Lower VMI datasets should be avoided due to larger deviations in measured value as compared to 70 keV, a level considered equivalent to conventional 120 kVp exams.

Available only for authorised users

Pickhardt PJ (2022) Value-added Opportunistic CT Screening: State of the Art. Radiology 303:241–254. https://doi.org/10.1148/radiol.211561CrossRefPubMed

Pickhardt PJ, Graffy PM, Perez AA, et al (2021) Opportunistic Screening at Abdominal CT: Use of Automated Body Composition Biomarkers for Added Cardiometabolic Value. RadioGraphics 41:524–542. https://doi.org/10.1148/rg.2021200056CrossRefPubMed

Lee MH, Zea R, Garrett JW, et al (2023) Abdominal CT Body Composition Thresholds Using Automated AI Tools for Predicting 10-year Adverse Outcomes. Radiology 306:e220574. https://doi.org/10.1148/radiol.220574CrossRefPubMed

Pickhardt PJ, Graffy PM, Zea R, et al (2020) Automated CT biomarkers for opportunistic prediction of future cardiovascular events and mortality in an asymptomatic screening population: a retrospective cohort study. Lancet Digit Health 2:e192–e200. https://doi.org/10.1016/S2589-7500(20)30025-XCrossRefPubMedPubMedCentral

Cheng X, Zhao K, Zha X, et al (2021) Opportunistic Screening Using Low-Dose CT and the Prevalence of Osteoporosis in China: A Nationwide, Multicenter Study. J Bone Miner Res Off J Am Soc Bone Miner Res 36:427–435. https://doi.org/10.1002/jbmr.4187CrossRef

Prado CM, Cushen SJ, Orsso CE, Ryan AM (2016) Sarcopenia and cachexia in the era of obesity: clinical and nutritional impact. Proc Nutr Soc 75:188–198. https://doi.org/10.1017/S0029665115004279CrossRefPubMed

Szczykutowicz TP, Pomerantz S, Siegel E, et al (2022) CT Protocol Standardization for Radiology AI: Cutting the Gordian Knot of Clinical Implementation. Radiol Manage 44:

Daye D, Wiggins WF, Lungren MP, et al (2022) Implementation of Clinical Artificial Intelligence in Radiology: Who Decides and How? Radiology 305:555–563. https://doi.org/10.1148/radiol.212151CrossRefPubMed

Lee MH, Liu D, Garrett JW, et al (2024) Comparing fully automated AI body composition measures derived from thin and thick slice CT image data. Abdom Radiol 49:985–996. https://doi.org/10.1007/s00261-023-04135-1CrossRef

10.

Perez AA, Pickhardt PJ, Elton DC, et al (2021) Fully automated CT imaging biomarkers of bone, muscle, and fat: correcting for the effect of intravenous contrast. Abdom Radiol 46:1229–1235. https://doi.org/10.1007/s00261-020-02755-5CrossRef

11.

Pickhardt PJ, Lauder T, Pooler BD, et al (2016) Effect of IV contrast on lumbar trabecular attenuation at routine abdominal CT: correlation with DXA and implications for opportunistic osteoporosis screening. Osteoporos Int 27:147–152. https://doi.org/10.1007/s00198-015-3224-9CrossRefPubMed

12.

Toia GV, Kim S, Dighe MK, Mileto A (2018) Dual-Energy Computed Tomography in Body Imaging. Semin Roentgenol 53:132–146. https://doi.org/10.1053/j.ro.2018.02.004CrossRefPubMed

13.

Reimer RP, Große Hokamp N, Fehrmann Efferoth A, et al (2021) Virtual monoenergetic images from spectral detector computed tomography facilitate washout assessment in arterially hyper-enhancing liver lesions. Eur Radiol 31:3468–3477. https://doi.org/10.1007/s00330-020-07379-3CrossRefPubMed

14.

Hokamp NG, Gilkeson R, Jordan MK, et al (2019) Virtual monoenergetic images from spectral detector CT as a surrogate for conventional CT images: Unaltered attenuation characteristics with reduced image noise. Eur J Radiol 117:49–55. https://doi.org/10.1016/j.ejrad.2019.05.019CrossRef

15.

Kordbacheh H, Baliyan V, Singh P, et al (2019) Rapid kVp switching dual-energy CT in the assessment of urolithiasis in patients with large body habitus: preliminary observations on image quality and stone characterization. Abdom Radiol N Y 44:1019–1026. https://doi.org/10.1007/s00261-018-1808-5CrossRef

16.

Baliyan V, Kordbacheh H, Pourvaziri A, et al (2020) Rapid kVp-switching DECT portal venous phase abdominal CT scans in patients with large body habitus: image quality considerations. Abdom Radiol N Y 45:2902–2909. https://doi.org/10.1007/s00261-020-02416-7CrossRef

17.

Szczykutowicz TP, Viggiano B, Rose S, et al (2021) A Metric for Quantification of Iodine Contrast Enhancement (Q-ICE) in Computed Tomography. J Comput Assist Tomogr 45:870. https://doi.org/10.1097/RCT.0000000000001215CrossRefPubMed

18.

Nikolau EP, Toia GV, Nett B, et al (2023) A Characterization of Deep Learning Reconstruction Applied to Dual-Energy Computed Tomography Monochromatic and Material Basis Images. J Comput Assist Tomogr 47:437–444. https://doi.org/10.1097/RCT.0000000000001442CrossRefPubMed

19.

Rassouli N, Chalian H, Rajiah P, et al (2017) Assessment of 70-keV virtual monoenergetic spectral images in abdominal CT imaging: A comparison study to conventional polychromatic 120-kVp images. Abdom Radiol 42:2579–2586. https://doi.org/10.1007/s00261-017-1151-2CrossRef

20.

Krishna S, Sadoughi N, McInnes MDF, et al (2018) Attenuation and Degree of Enhancement With Conventional 120-kVp Polychromatic CT and 70-keV Monochromatic Rapid Kilovoltage-Switching Dual-Energy CT in Cystic and Solid Renal Masses. Am J Roentgenol 211:789–796. https://doi.org/10.2214/AJR.17.19226CrossRef

21.

Darras KE, McLaughlin PD, Kang H, et al (2016) Virtual monoenergetic reconstruction of contrast-enhanced dual energy CT at 70 keV maximizes mural enhancement in acute small bowel obstruction. Eur J Radiol 85:950–956. https://doi.org/10.1016/j.ejrad.2016.02.019CrossRefPubMed

22.

Noda Y, Goshima S, Nakashima Y, et al (2020) Iodine dose optimization in portal venous phase virtual monochromatic images of the abdomen: Prospective study on rapid kVp switching dual energy CT. Eur J Radiol 122:108746. https://doi.org/10.1016/j.ejrad.2019.108746CrossRefPubMed

23.

Graffy PM, Sandfort V, Summers RM, Pickhardt PJ (2019) Automated Liver Fat Quantification at Nonenhanced Abdominal CT for Population-based Steatosis Assessment. Radiology 293:334–342. https://doi.org/10.1148/radiol.2019190512CrossRefPubMed

24.

Lee SJ, Liu J, Yao J, et al (2018) Fully automated segmentation and quantification of visceral and subcutaneous fat at abdominal CT: application to a longitudinal adult screening cohort. Br J Radiol 91:20170968. https://doi.org/10.1259/bjr.20170968CrossRefPubMedPubMedCentral

25.

Çiçek Ö, Abdulkadir A, Lienkamp SS, et al (2016) 3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation. In: Ourselin S, Joskowicz L, Sabuncu MR, et al (eds) Medical Image Computing and Computer-Assisted Intervention– MICCAI 2016. Springer International Publishing, Cham, pp 424–432

26.

Graffy PM, Liu J, Pickhardt PJ, et al (2019) Deep learning-based muscle segmentation and quantification at abdominal CT: application to a longitudinal adult screening cohort for sarcopenia assessment. Br J Radiol 92:20190327. https://doi.org/10.1259/bjr.20190327CrossRefPubMedPubMedCentral

27.

Jia Y, Shelhamer E, Donahue J, et al (2014) Caffe: Convolutional Architecture for Fast Feature Embedding. In: Proceedings of the 22nd ACM international conference on Multimedia. Association for Computing Machinery, New York, NY, USA, pp 675–678

28.

Sandfort V, Yan K, Pickhardt PJ, Summers RM (2019) Data augmentation using generative adversarial networks (CycleGAN) to improve generalizability in CT segmentation tasks. Sci Rep 9:16884. https://doi.org/10.1038/s41598-019-52737-xCrossRefPubMedPubMedCentral

29.

Summers RM, Elton DC, Lee S, et al (2021) Atherosclerotic Plaque Burden on Abdominal CT: Automated Assessment With Deep Learning on Noncontrast and Contrast-enhanced Scans. Acad Radiol 28:1491–1499. https://doi.org/10.1016/j.acra.2020.08.022CrossRefPubMed

30.

Ronneberger O, Fischer P, Brox T (2015) U-Net: Convolutional Networks for Biomedical Image Segmentation. In: Navab N, Hornegger J, Wells WM, Frangi AF (eds) Medical Image Computing and Computer-Assisted Intervention– MICCAI 2015. Springer International Publishing, Cham, pp 234–241

31.

Iglovikov V, Shvets A (2018) TernausNet: U-Net with VGG11 Encoder Pre-Trained on ImageNet for Image Segmentation

32.

Pickhardt PJ, Nguyen T, Perez AA, et al (2022) Improved CT-based Osteoporosis Assessment with a Fully Automated Deep Learning Tool. Radiol Artif Intell 4:e220042. https://doi.org/10.1148/ryai.220042CrossRefPubMedPubMedCentral

33.

Buslaev A, Parinov A, Khvedchenya E, et al (2020) Albumentations: fast and flexible image augmentations. Information 11:125. https://doi.org/10.3390/info11020125CrossRef

34.

Kingma DP, Ba J (2017) Adam: A Method for Stochastic Optimization

35.

36.

Yan K, Lu L, Summers RM (2018) Unsupervised body part regression via spatially self-ordering convolutional neural networks. In: 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018). pp 1022–1025

37.

Toia GV, Mileto A, Wang CL, Sahani DV (2022) Quantitative dual-energy CT techniques in the abdomen. Abdom Radiol N Y 47:3003–3018. https://doi.org/10.1007/s00261-021-03266-7CrossRef

38.

Rubino JG, Nasirzadeh AR, van der Pol CB, et al (2022) Quantitative and qualitative liver CT: imaging feature association with histopathologically confirmed hepatic cirrhosis. Abdom Radiol N Y 47:2314–2324. https://doi.org/10.1007/s00261-022-03550-0CrossRef

39.

Pham JT, Kalantari J, Ji C, et al (2020) Quantitative CT Predictors of Portal Venous Intervention in Uncontrolled Variceal Bleeding. AJR Am J Roentgenol 215:1247–1251. https://doi.org/10.2214/AJR.19.22460CrossRefPubMed

40.

Liu D, Garrett JW, Lee MH, et al (2023) Fully automated CT-based adiposity assessment: comparison of the L1 and L3 vertebral levels for opportunistic prediction. Abdom Radiol N Y 48:787–795. https://doi.org/10.1007/s00261-022-03728-6CrossRef

41.

Zatz LM, Alvarez RE (1977) An Inaccuracy in Computed Tomography: The Energy Dependence of CT Values. Radiology 124:91–97. https://doi.org/10.1148/124.1.91CrossRefPubMed

42.

Millner MR, Payne WH, Waggener RG, et al (1978) Determination of effective energies in CT calibration. Med Phys 5:543–545. https://doi.org/10.1118/1.594488CrossRefPubMed

43.

Schramm P, Huang Y, Erb G, et al (2009) How does the injection protocol influence the attenuation-time curve in CT perfusion measurements: Comparison of measured and simulated data. Med Phys 36:3487–3494. https://doi.org/10.1118/1.3159034CrossRefPubMed

44.

Li J, Udayasankar UK, Tang X, et al (2010) An Optimal Contrast Dose Indicator for the Determination of Hepatic Enhancement in Abdominal Multidetector Computed Tomography: Comparison of Patient Attenuation Indicator With Total Body Weight and Body Mass Index. J Comput Assist Tomogr 34:874. https://doi.org/10.1097/RCT.0b013e3181ed2f72CrossRefPubMed

45.

Kim EY, Yeh DW, Choe YH, et al (2010) Image quality and attenuation values of multidetector CT coronary angiography using high iodine-concentration contrast material: a comparison of the use of iopromide 370 and iomeprol 400. Acta Radiol 51:982–989. https://doi.org/10.3109/02841851.2010.509740CrossRefPubMedPubMedCentral

46.

Lortie J, Gage G, Rush B, et al (2022) The effect of computed tomography parameters on sarcopenia and myosteatosis assessment: a scoping review. J Cachexia Sarcopenia Muscle 13:2807–2819. https://doi.org/10.1002/jcsm.13068CrossRefPubMedPubMedCentral

47.

Salyapongse AM, Rose SD, Pickhardt PJ, et al (2023) CT Number Accuracy and Association With Object Size: A Phantom Study Comparing Energy-Integrating Detector CT and Deep Silicon Photon-Counting Detector CT. AJR Am J Roentgenol 221:539–547. https://doi.org/10.2214/AJR.23.29463CrossRefPubMed

48.

Szczykutowicz TP, DuPlissis A, Pickhardt PJ (2017) Variation in CT Number and Image Noise Uniformity According to Patient Positioning in MDCT. AJR Am J Roentgenol 208:1064–1072. https://doi.org/10.2214/AJR.16.17215CrossRefPubMed

49.

Tharmaseelan H, Rotkopf LT, Ayx I, et al (2022) Evaluation of radiomics feature stability in abdominal monoenergetic photon counting CT reconstructions. Sci Rep 12:19594. https://doi.org/10.1038/s41598-022-22877-8CrossRefPubMedPubMedCentral

50.

Park S, Lee SM, Kim S, et al (2021) Performance of radiomics models for survival prediction in non-small-cell lung cancer: influence of CT slice thickness. Eur Radiol 31:2856–2865. https://doi.org/10.1007/s00330-020-07423-2CrossRefPubMed

51.

Lewis M, Reid K, Toms AP (2013) Reducing the effects of metal artefact using high keV monoenergetic reconstruction of dual energy CT (DECT) in hip replacements. Skeletal Radiol 42:275–282. https://doi.org/10.1007/s00256-012-1458-6CrossRefPubMed

52.

Neuhaus V, Große Hokamp N, Abdullayev N, et al (2017) Metal artifact reduction by dual-layer computed tomography using virtual monoenergetic images. Eur J Radiol 93:143–148. https://doi.org/10.1016/j.ejrad.2017.05.013CrossRefPubMed

53.

Cesareo R, Hanson AL, Gigante GE, et al (1992) Interaction of keV photons with matter and new applications. Phys Rep 213:117–178. https://doi.org/10.1016/0370-1573(92)90086-FCrossRef

54.

Starekova J, Hernando D, Pickhardt PJ, Reeder SB (2021) Quantification of Liver Fat Content with CT and MRI: State of the Art. Radiology 301:250–262. https://doi.org/10.1148/radiol.2021204288CrossRefPubMed

55.

Garner HW, Paturzo MM, Gaudier G, et al (2017) Variation in Attenuation in L1 Trabecular Bone at Different Tube Voltages: Caution Is Warranted When Screening for Osteoporosis With the Use of Opportunistic CT. Am J Roentgenol 208:165–170. https://doi.org/10.2214/AJR.16.16744CrossRef

56.

Pooler BD, Garrett JW, Southard AM, et al (2023) Technical Adequacy of Fully Automated Artificial Intelligence Body Composition Tools: Assessment in a Heterogeneous Sample of External CT Examinations. Am J Roentgenol 221:1–9. https://doi.org/10.2214/AJR.22.28745CrossRef

57.

Kalisz K, Rassouli N, Dhanantwari A, et al (2018) Noise characteristics of virtual monoenergetic images from a novel detector-based spectral CT scanner. Eur J Radiol 98:118–125. https://doi.org/10.1016/j.ejrad.2017.11.005CrossRefPubMed

Title: Comparing fully automated AI body composition biomarkers at differing virtual monoenergetic levels using dual-energy CT
Authors: Giuseppe V. Toia
John W. Garret
Sean D. Rose
Timothy P. Szczykutowicz
Perry J. Pickhardt
Publication date: 07-12-2024
Publisher: Springer US
Keywords: Biomarkers
Artificial Intelligence
Published in: Abdominal Radiology
Print ISSN: 2366-004X
Electronic ISSN: 2366-0058
DOI: https://doi.org/10.1007/s00261-024-04733-7

Elevate your expertise in aplastic anemia (Link opens in a new window)

Aplastic Anemia
Independent Medical Education

Transform the way you care for your patients with aplastic anemia with our 3-module series using real-world case studies and expert insights. Discover why early diagnosis matters, explore the benefits and risks of current treatments, and develop tailored approaches for complex cases.

Supported by:

Pfizer

Developed by: Springer Healthcare IME

Learn more

Keynote series | Spotlight on menopause

Menopause
Video Series

Menopause can have a significant impact on the body, with effects ranging beyond the endocrine and reproductive systems. Learn about the systemic effects of menopause, so you can help patients in your clinics through the transition.

Prof. Martha Hickey

Dr. Claudia Barth

Dr. Samar El Khoudary

Developed by: Springer Medicine

Watch now