Abstract
Lower limb muscles provide various significant physiological functions, which are the key muscles to support daily activities. Segmentation of these muscles offers quantitative analysis for several clinical applications (e.g., understanding of musculoskeletal diseases). In recent studies on muscle tissues, water-fat quantitative magnetic resonance imaging (MRI) is frequently used as the main image modality due to its capability in distinguishing muscle and fat from their surrounding tissues. However, manual muscle segmentation in MRI is time-consuming and often requires professional knowledge. In this paper, an ensemble framework of combining two patch-based binary-class deep convolutional neural networks with the same architecture is proposed to achieve thigh and calf muscle segmentation from whole-body water-fat (mDIXON) MRIs. We compared our model to a state-of-the-art multi-class 3D U-Net model using a 5-fold cross-validation. A Dice coefficient of 0.9042 was achieved by our method, which was significantly more accurate than the multi-class 3D U-Net method. Additionally, the model deployment costs approximately 13.8 s per case.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Hamrick, M.W., McGee-Lawrence, M.E., Frechette, D.M.: Fatty infiltration of skeletal muscle: mechanisms and comparisons with bone marrow adiposity. Front. Endocrinol. 7, 69 (2016)
Ding, J., Cao, P., Chang, H.C., Gao, Y., Chan, S.H.S., Vardhanabhuti, V.: Deep learning-based thigh muscle segmentation for reproducible fat fraction quantification using fat–water decomposition MRI. Insights Imaging 11(1), 1–11 (2020)
Ronneberger, O., Fischer, P., Brox, T.: U-net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-24574-4_28
Amer, R., Nassar, J., Bendahan, D., Greenspan, H., Ben-Eliezer, N.: Automatic segmentation of muscle tissue and inter-muscular fat in thigh and calf MRI images. In: Shen, D., et al. (eds.) MICCAI 2019. LNCS, vol. 11765, pp. 219–227. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-32245-8_25
Ahmad, E., Goyal, M., McPhee, J.S., Degens, H., Yap, M.H.: Semantic segmentation of human thigh quadriceps muscle in magnetic resonance images. arXiv preprint arXiv:1801.00415 (2018)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems 25, pp. 1097–1105 (2012)
Long, J., Shelhamer, E., Darrell, T.: Fully convolutional networks for semantic segmentation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3431–3440 (2015)
Zhao, H., Shi, J., Qi, X., Wang, X., Jia, J.: Pyramid scene parsing network. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2881–2890 (2017)
Ghosh, S., Boulanger, P., Acton, S.T., Blemker, S.S., Ray, N.: Automated 3D muscle segmentation from MRI data using convolutional neural network. In: 2017 IEEE International Conference on Image Processing (ICIP), pp. 4437–4441. IEEE, September 2017
Ni, R., Meyer, C.H., Blemker, S.S., Hart, J.M., Feng, X.: Automatic segmentation of all lower limb muscles from high-resolution magnetic resonance imaging using a cascaded three-dimensional deep convolutional neural network. J. Med. Imaging 6(4), 044009 (2019)
Çiçek, Ö., Abdulkadir, A., Lienkamp, S.S., Brox, T., Ronneberger, O.: 3D U-net: learning dense volumetric segmentation from sparse annotation. In: Ourselin, S., Joskowicz, L., Sabuncu, M.R., Unal, G., Wells, W. (eds.) MICCAI 2016. LNCS, vol. 9901, pp. 424–432. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46723-8_49
Zhang, Y., Brady, M., Smith, S.: Segmentation of brain MR images through a hidden Markov random field model and the expectation-maximization algorithm. IEEE Trans. Med. Imaging 20(1), 45–57 (2001)
Dice, L.R.: Measures of the amount of ecologic association between species. Ecology 26(3), 297–302 (1945)
Altman, D.G., Bland, J.M.: Measurement in medicine: the analysis of method comparison studies. J. R. Stat. Soc. Ser. D (Stat.) 32(3), 307–317 (1983)
Acknowledgement
This research was funded by the NIHR Nottingham Biomedical Research Centre and carried out at/supported by the NIHR Nottingham Clinical Research Facilities. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health and Social Care. Funding was also provided by the Centre for musculoskeletal Ageing Research, University of Nottingham to support the PhD of Rosemary Nicholas.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Gong, Z., Nicholas, R., Francis, S.T., Chen, X. (2022). Thigh and Calf Muscles Segmentation Using Ensemble of Patch-Based Deep Convolutional Neural Network on Whole-Body Water-Fat MRI. In: Yang, G., Aviles-Rivero, A., Roberts, M., Schönlieb, CB. (eds) Medical Image Understanding and Analysis. MIUA 2022. Lecture Notes in Computer Science, vol 13413. Springer, Cham. https://doi.org/10.1007/978-3-031-12053-4_20
Download citation
DOI: https://doi.org/10.1007/978-3-031-12053-4_20
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-12052-7
Online ISBN: 978-3-031-12053-4
eBook Packages: Computer ScienceComputer Science (R0)