Abstract
This paper concerns fully automatic and objective measurement of human skeletal muscle fiber orientation directly from standard b-mode ultrasound images using deep residual (ResNet) and convolutional neural networks (CNN). Fiber orientation and length is related with active and passive states of force production within muscle. There is currently no non-invasive way to measure force directly from muscle. Measurement of forces and other contractile parameters like muscle length change, thickness, and tendon length is not only important for understanding healthy muscle, but such information has contributed to understanding, diagnosis, monitoring, targeting and treatment of diseases ranging from myositis to stroke and motor neurone disease (MND). We applied well established deep learning methods to ultrasound data recorded from 19 healthy participants (5 female, ages: 30 ± 7.7) and achieved state of the art accuracy in predicting fiber orientation directly from ultrasound images of the calf muscles. First we used a previously developed segmentation technique to extract a region of interest within the gastrocnemius muscle. Then we asked an expert to annotate the main line of fiber orientation in \( 4 \times 4 \) partitions of 400 normalized images. A linear model was then applied to the annotations to regulate and recover the orientation field for each image. Then we applied a CNN and a ResNet to predict the fiber orientation in each image. With leave one participant out cross-validation and dropout as a regulariser, we were able to demonstrate state of the art performance, recovering the fiber orientation with an average error of just \( 2^\circ \).
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Barry, D., Ahmed, A.M.: Design and performance of a modified buckle transducer for the measurement of ligament tension. J. Biomech. Eng. 108(2), 149–152 (1986). https://doi.org/10.1115/1.3138594
Cunningham, R., Harding, P., Loram, I.: Real-time ultrasound segmentation, analysis and visualization of deep cervical muscle structure. Trans. Med. Imaging 36(2), 653–665 (2015). https://doi.org/10.1109/TMI.2016.2623819
Darby, J., Hodson-Tole, E.F., Costen, N., Loram, I.D.: Automated regional analysis of B-mode ultrasound images of skeletal muscle movement. J. Appl. Physiol. 112(2), 313–327 (2012). https://doi.org/10.1152/japplphysiol.00701.2011
Darby, J., Li, B., Costen, N., Loram, I., Hodson-Tole, E.: Estimating skeletal muscle fascicle curvature from B-mode ultrasound image sequences. IEEE Trans. Biomed. Eng. 60(7), 1935–1945 (2013). https://doi.org/10.1109/TBME.2013.2245328
Finni, T., Komi, P.V., Lukkariniemi, J.: Achilles tendon loading during walking: application of a novel optic fiber technique. Eur. J. Appl. Physiol. Occup. Physiol. 77(3), 289–291 (1998). https://doi.org/10.1007/s004210050335
Finni, T., Komi, P.V., Lepola, V.: In vivo human triceps surae and quadriceps femoris muscle function in a squat jump and counter. Eur. J. Appl. Physiol. 83, 416–426 (2000). https://doi.org/10.1007/s004210000289
Frangi, A.F., Niessen, W.J., Vincken, K.L., Viergever, M.A.: Multiscale vessel enhancement filtering. In: Wells, W.M., Colchester, A., Delp, S. (eds.) MICCAI 1998. LNCS, vol. 1496, pp. 130–137. Springer, Heidelberg (1998). doi:10.1007/BFb0056195. https://doi.org/10.1016/j.media.2004.08.001
Gregor, R.J., Komi, P.V., Järvinen, M.: Achilles tendon forces during cycling. Int. J. Sports Med. 8(Suppl. 1), 9–14 (1987). https://doi.org/10.1055/s-2008-1025698
Harding, P.J., Loram, I.D., Combes, N., Hodson-Tole, E.F.: Ultrasound-based detection of fasciculations in healthy and diseased muscles. IEEE Trans. Biomed. Eng. 63(3), 512–518 (2016). https://doi.org/10.1109/TBME.2015.2465168
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. Arxiv.Org 7(3), 171–180 (2015). https://doi.org/10.3389/fpsyg.2013.00124
Herbert, R.D., Gandevia, S.C., Herbert, R.D.: Changes in pennation with joint angle and muscle torque: in vivo measurements in human brachialis muscle. J. Physiol. 484(Pt2), 523–532 (1995). https://doi.org/10.1113/jphysiol.1995.sp020683
Hinton, G.: Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15, 1929–1958 (2014)
Holden, J.P., Grood, E.S., Korvick, D.L., Cummings, J.F., Butler, D.L., Bylski-Austrow, D.I.: In vivo forces in the anterior cruciate ligament: direct measurements during walking and trotting in a quadruped. J. Biomech. 27(5), 517–526 (1994). https://doi.org/10.1016/0021-9290(94)90063-9
Komi, P.V.: Optic fibre as a transducer of tendomuscular forces. Eur. J. Appl. Physiol. Occup. Physiol. 72(3), 278–280 (1996). https://doi.org/10.1007/BF00838652
Komi, P.V., Salonen, M., Jarvinen, M., Kokko, O.: In vivo registration of Achilles tendon forces in man. I. Methodological development. Int. J. Sports Med. 8(Suppl. 1), 3–8 (1987). https://doi.org/10.1055/s-2008-1025697
Krizhevsky, A., Sutskever, I., Hinton, G. E.: ImageNet classification with deep convolutional neural networks. Adv. Neural Inf. Process. Syst. 25, 1–9 (2012a). https://doi.org/10.1117/12.2176558
Krizhevsky, A., Sutskever, I., Hinton, G.: Imagenet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems, pp. 1097–1105 (2012b)
LeCun, Y., Boser, B., Denker, J.S., Henderson, D., Howard, R.E., Hubbard, W., Jackel, L.D.: Handwritten digit recognition with a back-propagation network. In: Advances in Neural Information Processing Systems, pp. 396–404 (1990). https://doi.org/10.1111/dsu.12130
Lewis, J.L., Lew, W.D., Schmidt, J.: A note on the application and evaluation of the buckle transducer for the knee ligament force measurement. J. Biomech. Eng. 104(2), 125–128 (1982). http://www.ncbi.nlm.nih.gov/pubmed/7078126
Lieber, R.L., Fridén, J.: Functional and clinical significance of skeletal muscle architecture. Muscle Nerve 23(11), 1647–1666 (2000). https://doi.org/10.1002/1097-4598(200011)23:11<1647:AID-MUS1>3.0.CO;2-M
Loram, I., Bate, B., Harding, P., Cunningham, R., Loram, A.: Proactive selective inhibition targeted at the neck muscles: this proximal constraint facilitates learning and regulates global control. IEEE Trans. Neural Syst. Rehabil. Eng. 25(4), 357–369 (2017). https://doi.org/10.1109/TNSRE.2016.2641024
Loram, I., Maganaris, C., Lakie, M.: Use of ultrasound to make noninvasive in vivo measurement of continuous changes in human muscle contractile length. J. Appl. Physiol. 100, 1311–1323 (2006). https://doi.org/10.1152/japplphysiol.01229.2005
Nair, V., Hinton, G.E.: Rectified linear units improve restricted Boltzmann machines. Proc. Int. Conf. Mach. Learn. 3, 807–814 (2010). https://doi.org/10.1.1.165.6419
Namburete, A.I.L., Rana, M., Wakeling, J.M.: Computational methods for quantifying in vivo muscle fascicle curvature from ultrasound images. J. Biomech. 44(14), 2538–2543 (2011). https://doi.org/10.1016/j.jbiomech.2011.07.017
Narici, M.V., Binzoni, T., Hiltbrand, E., Fasel, J., Terrier, F., Cerretelli, P.: In vivo human gastrocnemius architecture with changing joint angle at rest and during graded isometric contraction. J. Physiol. 496(1), 287–297 (1996). https://doi.org/10.1113/jphysiol.1996.sp021685
Rana, M., Hamarneh, G., Wakeling, J.M.: Automated tracking of muscle fascicle orientation in B-mode ultrasound images. J. Biomech. 42(13), 2068–2073 (2009). https://doi.org/10.1016/j.jbiomech.2009.06.003
Shi, J., Tomasi, C.: Good features to track. In: Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 1994, pp. 593–600 (1994). https://doi.org/10.1109/cvpr.1994.323794
Tomasi, C.: Detection and tracking of point features. School of Computer Science, Carnegie Mellon University, 91(April), 1–22 (1991). https://doi.org/10.1016/S0031-3203(03)00234-6
Yeung, F., Levinson, S.F., Parker, K.J.: Multilevel and motion model-based ultrasonic speckle tracking algorithms. Ultrasound Med. Biol. 24(3), 427–441 (1998). https://doi.org/10.1016/S0301-5629(97)00281-0
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Cunningham, R., Harding, P., Loram, I. (2017). Deep Residual Networks for Quantification of Muscle Fiber Orientation and Curvature from Ultrasound Images. In: Valdés Hernández, M., González-Castro, V. (eds) Medical Image Understanding and Analysis. MIUA 2017. Communications in Computer and Information Science, vol 723. Springer, Cham. https://doi.org/10.1007/978-3-319-60964-5_6
Download citation
DOI: https://doi.org/10.1007/978-3-319-60964-5_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-60963-8
Online ISBN: 978-3-319-60964-5
eBook Packages: Computer ScienceComputer Science (R0)