Automated segmentation of 2D low-dose CT images of the psoas-major muscle using deep convolutional neural networks

  • Fumio HashimotoEmail author
  • Akihiro Kakimoto
  • Nozomi Ota
  • Shigeru Ito
  • Sadahiko Nishizawa


The psoas-major muscle has been reported as a predictive factor of sarcopenia. The cross-sectional area (CSA) of the psoas-major muscle in axial images has been indicated to correlate well with the whole-body skeletal muscle mass. In this study, we evaluated the segmentation accuracy of low-dose X-ray computed tomography (CT) images of the psoas-major muscle using the U-Net convolutional neural network, which is a deep-learning technique. Deep learning has been recently known to outperform conventional image-segmentation techniques. We used fivefold cross validation to validate the segmentation performance (n = 100) of the psoas-major muscle. For the intersection over union and CSA ratio, segmentation accuracies of 86.0 and 103.1%, respectively, were achieved. These results suggest that the U-Net network is competitive compared with the previous methods. Therefore, the proposed technique is useful for segmenting the psoas-major muscle even in low-dose CT images.


Psoas-major muscle Deep learning Convolutional neural networks X-ray computed tomography Automated segmentation Sarcopenia 



We would like to thank the staff of the Hamamatsu Medical Imaging Center and Hamamatsu Photonics K. K. for their support.

Compliance with ethical standards

Conflict of interest

We declare that this work is free from financial limitation or any other relationship that might lead to a conflict of interest. The authors have no conflicts of interest to declare.

Statement of human and animal rights

All procedures performed in the studies involving human participants were in accordance with the ethical standards of the Institutional Review Board of Hamamatsu Medical Photonics Foundation and 1964 Helsinki declaration and its later amendments or comparable ethical standards. This article does not contain any studies performed with animals.

Informed consent

Informed consent was obtained from all study participants.


  1. 1.
    Kim JD, Kuno S, Soma R, Masuda K, et al. Relationship between reduction of hip joint and thigh muscle and walking ability in elderly people. Jpn J Phys Fit Sports Med. 2000;49(5):737–8. Scholar
  2. 2.
    Yaguchi Y, Kumata Y, Horikawa M, et al. Clinical significance of area of psoas major muscle on computed tomography after gastrectomy in gastric cancer patients. Ann Nutr Metab. 2017;71(3–4):145–9.CrossRefGoogle Scholar
  3. 3.
    Fearon K, Strasser F, Anker SD, et al. Definition and classification of cancer cachexia: an international consensus. Lancet Oncol. 2011;12(5):489–95.CrossRefGoogle Scholar
  4. 4.
    Shen W, Punyanitya M, Wang Z, et al. Total body skeletal muscle and adipose tissue volumes: estimation from a single abdominal cross-sectional image. J Appl Physiol. 2004;97(6):2333–8.CrossRefGoogle Scholar
  5. 5.
    Baracos VE, Reiman T, Mourtzakis M, Gioulbasanis I, Antoun S. Body composition in patients with non-small cell lung cancer: a contemporary view of cancer cachexia with the use of computed tomography image analysis. Am J Clin Nutr. 2010;91(4):1133–7.CrossRefGoogle Scholar
  6. 6.
    Martin L, Birdsell L, MacDonald N, et al. Cancer cachexia in the age of obesity: skeletal muscle depletion is a powerful prognostic factor, independent of body mass index. J Clin Oncol. 2013;31(12):1539–47.CrossRefGoogle Scholar
  7. 7.
    Meesters SPL, Yokota F, Okada T, et al. Multi atlas-based muscle segmentation in abdominal CT images with varying field of view. Paper presented at the International Forum on Medical Imaging in Asia (IFMIA), November 16–17, 2012, Daejon Korea.Google Scholar
  8. 8.
    Kamiya N, Zhou X, Chen H, et al. Automated segmentation of psoas major muscle in X-ray CT images by use of a shape model: preliminary study. Radiol Phys Technol. 2012;5(1):5–14.CrossRefGoogle Scholar
  9. 9.
    Inoue T, Kitamura Y, Li Y, Ito W, Ishikawa H. Psoas major muscle segmentation using higher-order shape prior. In: International MICCAI workshop on medical computer vision, 2016; p. 116–24.
  10. 10.
    Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation. IEEE Trans Pattern Anal Mach Intell. 2016;39(4):640–51.Google Scholar
  11. 11.
    Badrinarayanan V, Kendall A, Cipolla R. Segnet: a deep convolutional encoder-decoder architecture for image segmentation. IEEE Trans Pattern Anal Mach Intell. 2017;39(12):2481–95.CrossRefGoogle Scholar
  12. 12.
    Ronneberger O, Fischer P, Brox T. U-net: convolutional networks for biomedical image segmentation. In: International conference on medical image computing and computer-assisted intervention 2015; p. 234–41.
  13. 13.
    Zhao H, Shi J, Qi X, Wang X, Jia J. Pyramid scene parsing network. In: IEEE conference on computer vision and pattern recognition (CVPR) 2017; p. 2881–90.
  14. 14.
    Greenspan H, van Ginneken B, Summers RM. Guest editorial deep learning in medical imaging: overview and future promise of an exciting new technique. IEEE Trans Med Imaging. 2016;35(5):1153–9.CrossRefGoogle Scholar
  15. 15.
    Suzuki K. Overview of deep learning in medical imaging. Radiol Phys Technol. 2017;10(3):257–73.CrossRefGoogle Scholar
  16. 16.
    Teramoto A, Tsukamoto T, Kiriyama Y, Fujita H. Automated classification of lung cancer types from cytological images using deep convolutional neural networks. Biomed Res Int. 2017;4067832:1–6.CrossRefGoogle Scholar
  17. 17.
    Zhou X, Takayama R, Wang S, Hara T, Fujita H. Deep learning of the sectional appearances of 3D CT images for anatomical structure segmentation based on an FCN voting method. Med Phys. 2017;44(10):5221–33.CrossRefGoogle Scholar
  18. 18.
    Lee H, Troschel FM, Tajmir S, et al. Pixel-level deep segmentation: artificial intelligence quantifies muscle on computed tomography for body morphometric analysis. J Digit Imaging. 2017;30(4):487–98.CrossRefGoogle Scholar
  19. 19.
    Ghosh S, Boulanger P, Acton ST, Blemker SS, Ray N. Automated 3D muscle segmentation from MRI data using convolutional neural network. In: IEEE international conference on image processing (ICIP) 2017; p. 4437–41.
  20. 20.
    Kamiya N, Kume M, Zheng G, et al. Automated recognition of erector spinae muscles and their skeletal attachment region via deep learning in torso CT images. In: International workshop on computational methods and clinical applications in musculoskeletal imaging. 2018.
  21. 21.
    Abadi M, Barham P, Chen J, et al. TensorFlow: a system for large-scale machine learning. In: Proc 12th USENIX conf operating syst design implement. 2016;16:265–83.Google Scholar
  22. 22.
    Keras: The Python Deep Learning library. Accessed 30 Mar 2019.
  23. 23.
    Kingma DP, Ba J. Adam: a method for stochastic optimization. arXiv:1412.6980.
  24. 24.
    Mahendran A, Vedaldi A. Understanding deep image representations by inverting them. In: IEEE international conference on computer vision and pattern recognition (CVPR) 2014; p. 5188–96.
  25. 25.
    Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D. Grad-cam: visual explanations from deep networks via gradient-based localization. In: IEEE international conference on computer vision (ICCV) 2017; p. 618–26.

Copyright information

© Japanese Society of Radiological Technology and Japan Society of Medical Physics 2019

Authors and Affiliations

  1. 1.Central Research LaboratoryHamamatsu Photonics K.K.HamamatsuJapan
  2. 2.Global Strategic Challenge CenterHamamatsu Photonics K.K.HamamatsuJapan
  3. 3.Hamamatsu Medical Imaging Center, Hamamatsu Medical Photonics FoundationHamamatsuJapan

Personalised recommendations