Abstract
Purpose
In quantitative computed tomography (CT), manual selection of the intensity calibration phantom’s region of interest is necessary for calculating density (mg/cm3) from the radiodensity values (Hounsfield units: HU). However, as this manual process requires effort and time, the purposes of this study were to develop a system that applies a convolutional neural network (CNN) to automatically segment intensity calibration phantom regions in CT images and to test the system in a large cohort to evaluate its robustness.
Methods
This cross-sectional, retrospective study included 1040 cases (520 each from two institutions) in which an intensity calibration phantom (B-MAS200, Kyoto Kagaku, Kyoto, Japan) was used. A training dataset was created by manually segmenting the phantom regions for 40 cases (20 cases for each institution). The CNN model’s segmentation accuracy was assessed with the Dice coefficient, and the average symmetric surface distance was assessed through fourfold cross-validation. Further, absolute difference of HU was compared between manually and automatically segmented regions. The system was tested on the remaining 1000 cases. For each institution, linear regression was applied to calculate the correlation coefficients between HU and phantom density.
Results
The source code and the model used for phantom segmentation can be accessed at https://github.com/keisuke-uemura/CT-Intensity-Calibration-Phantom-Segmentation. The median Dice coefficient was 0.977, and the median average symmetric surface distance was 0.116 mm. The median absolute difference of the segmented regions between manual and automated segmentation was 0.114 HU. For the test cases, the median correlation coefficients were 0.9998 and 0.999 for the two institutions, with a minimum value of 0.9863.
Conclusion
The proposed CNN model successfully segmented the calibration phantom regions in CT images with excellent accuracy.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Data availability
The model used for phantom segmentation can be accessed via https://github.com/keisuke-uemura/CT-Intensity-Calibration-Phantom-Segmentation
Code availability
The code used for phantom segmentation can be accessed via https://github.com/keisuke-uemura/CT-Intensity-Calibration-Phantom-Segmentation
References
Kanis JA, Cooper C, Rizzoli R, Reginster JY (2019) European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Osteoporos Int 30(1):3–44. https://doi.org/10.1007/s00198-018-4704-5
Orimo H, Nakamura T, Hosoi T, Iki M, Uenishi K, Endo N, Ohta H, Shiraki M, Sugimoto T, Suzuki T, Soen S, Nishizawa Y, Hagino H, Fukunaga M, Fujiwara S (2012) Japanese 2011 guidelines for prevention and treatment of osteoporosis–executive summary. Arch Osteoporos 7(1–2):3–20. https://doi.org/10.1007/s11657-012-0109-9
Camacho PM, Petak SM, Binkley N, Diab DL, Eldeiry LS, Farooki A, Harris ST, Hurley DL, Kelly J, Lewiecki EM, Pessah-Pollack R, McClung M, Wimalawansa SJ, Watts NB (2020) American Association of clinical endocrinologists/American college of endocrinology clinical practice guidelines for the diagnosis and treatment of postmenopausal osteoporosis-2020 update. Endocr Pract Off J Am Coll Endocr Am Assoc Clin Endocr 26(Suppl 1):1–46. https://doi.org/10.4158/gl-2020-0524suppl
Maeda Y, Sugano N, Saito M, Yonenobu K (2011) Comparison of femoral morphology and bone mineral density between femoral neck fractures and trochanteric fractures. Clin Orthop Relat Res 469(3):884–889. https://doi.org/10.1007/s11999-010-1529-8
Uemura K, Takao M, Otake Y, Hamada H, Sakai T, Sato Y, Sugano N (2018) The distribution of bone mineral density in the femoral heads of unstable intertrochanteric fractures. J Orthop Surg 26(2):2309499018778325. https://doi.org/10.1177/2309499018778325
Whitmarsh T, Otake Y, Uemura K, Takao M, Sugano N, Sato Y (2019) A cross-sectional study on the age-related cortical and trabecular bone changes at the femoral head in elderly female hip fracture patients. Sci Rep 9(1):305. https://doi.org/10.1038/s41598-018-36299-y
Hanusch BC, Tuck SP, Mekkayil B, Shawgi M, McNally RJQ, Walker J, Francis RM, Datta HK (2020) Quantitative computed tomography (QCT) of the distal forearm in men using a spiral whole-body CT scanner: description of a method and reliability assessment of the QCT Pro software. J Clin Densitom Off J Int Soc Clin Densitom 23(3):418–425. https://doi.org/10.1016/j.jocd.2019.05.005
Adams JE (2009) Quantitative computed tomography. Eur J Radiol 71(3):415–424. https://doi.org/10.1016/j.ejrad.2009.04.074
Giambini H, Dragomir-Daescu D, Huddleston PM, Camp JJ, An KN, Nassr A (2015) The effect of quantitative computed tomography acquisition protocols on bone mineral density estimation. J Biomech Eng 137(11):114502. https://doi.org/10.1115/1.4031572
Lee DC, Hoffmann PF, Kopperdahl DL, Keaveny TM (2017) Phantomless calibration of CT scans for measurement of BMD and bone strength-Inter-operator reanalysis precision. Bone 103:325–333. https://doi.org/10.1016/j.bone.2017.07.029
Hiasa Y, Otake Y, Takao M, Ogawa T, Sugano N, Sato Y (2020) Automated muscle segmentation from clinical CT using Bayesian U-net for personalized musculoskeletal modeling. IEEE Trans Med Imaging 39(4):1030–1040. https://doi.org/10.1109/tmi.2019.2940555
He K, Zhang X, Ren S, Sun J (2015) Delving deep into rectifiers: surpassing human-level performance on imagenet classification. arXiv:150201852
Kingma DP, J B (2017) Adam: a method for stochastic optimization. arXiv:14126980
Dice LR (1945) Measures of the amount of ecologic association between species. Ecology 26(3):297–302. https://doi.org/10.2307/1932409
Styner M, Lee J, Chin B, Chin M, Commowick O, Tran H, Markovic-Plese S, Jewells V, Warfield S (2008) 3D segmentation in the clinic: a grand challenge II: MS lesion segmentation. Midas J 1–5
Aamodt A, Kvistad KA, Andersen E, Lund-Larsen J, Eine J, Benum P, Husby OS (1999) Determination of Hounsfield value for CT-based design of custom femoral stems. J Bone Joint surg Br 81(1):143–147
Gausden EB, Nwachukwu BU, Schreiber JJ, Lorich DG, Lane JM (2017) Opportunistic use of CT imaging for osteoporosis screening and bone density assessment: a qualitative systematic review. J Bone Joint Surg Am 99(18):1580–1590. https://doi.org/10.2106/jbjs.16.00749
Kitamura K, Fujii M, Utsunomiya T, Iwamoto M, Ikemura S, Hamai S, Motomura G, Todo M, Nakashima Y (2020) Effect of sagittal pelvic tilt on joint stress distribution in hip dysplasia: a finite element analysis. Clin Biomech 74:34–41. https://doi.org/10.1016/j.clinbiomech.2020.02.011
Schreiber JJ, Anderson PA, Rosas HG, Buchholz AL, Au AG (2011) Hounsfield units for assessing bone mineral density and strength: a tool for osteoporosis management. J Bone Joint Surg Am 93(11):1057–1063. https://doi.org/10.2106/jbjs.j.00160
Mawatari T, Hayashida Y, Katsuragawa S, Yoshimatsu Y, Hamamura T, Anai K, Ueno M, Yamaga S, Ueda I, Terasawa T, Fujisaki A, Chihara C, Miyagi T, Aoki T, Korogi Y (2020) The effect of deep convolutional neural networks on radiologists’ performance in the detection of hip fractures on digital pelvic radiographs. Eur J Radiol 130:109188. https://doi.org/10.1016/j.ejrad.2020.109188
Cheng CT, Ho TY, Lee TY, Chang CC, Chou CC, Chen CC, Chung IF, Liao CH (2019) Application of a deep learning algorithm for detection and visualization of hip fractures on plain pelvic radiographs. Eur Radiol 29(10):5469–5477. https://doi.org/10.1007/s00330-019-06167-y
Therkildsen J, Thygesen J, Winther S, Svensson M, Hauge EM, Böttcher M, Ivarsen P, Jørgensen HS (2018) Vertebral bone mineral density measured by quantitative computed tomography with and without a calibration phantom: a comparison between 2 different software solutions. J Clin Densitom Off J Int Soc Clin Densitom 21(3):367–374. https://doi.org/10.1016/j.jocd.2017.12.003
Feit A, Levin N, McNamara EA, Sinha P, Whittaker LG, Malabanan AO, Rosen HN (2019) Effect of positioning of the region of interest on bone density of the hip. J Clin Densitom Off J Int Soc Clin Densitom. https://doi.org/10.1016/j.jocd.2019.04.002
Acknowledgements
This study was supported by the Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research (KAKENHI) numbers 19H01176 and 20H04550. The authors thank Tatsuya Kitaura MD and the radiological technologists for their help with data acquisition.
Funding
This study was supported by the Japan Society for the Promotion of Science (JSPS) Grants-in-Aid for Scientific Research (KAKENHI) Numbers 19H01176 and 20H04550.
Author information
Authors and Affiliations
Contributions
KU and YO contributed to conceptualization and methodology; KU, YO, and MS were involved in code writing; KU and AK contributed to formal analysis and investigation; KU contributed to writing—original draft preparation; YO, MT, MS, NS, and YS contributed to writing—review and editing; YO and YS contributed to funding acquisition. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Conflict of interest
The authors have nothing to disclose.
Ethics approval
All procedures performed in this study were performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.
Consent to participate
This study was approved by the Institutional Review Board of each participating hospital, and written informed consent was waived because of the retrospective design.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Uemura, K., Otake, Y., Takao, M. et al. Automated segmentation of an intensity calibration phantom in clinical CT images using a convolutional neural network. Int J CARS 16, 1855–1864 (2021). https://doi.org/10.1007/s11548-021-02345-w
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11548-021-02345-w