Automated muscle segmentation from CT images of the hip and thigh using a hierarchical multi-atlas method
Patient-specific quantitative assessments of muscle mass and biomechanical musculoskeletal simulations require segmentation of the muscles from medical images. The objective of this work is to automate muscle segmentation from CT data of the hip and thigh.
We propose a hierarchical multi-atlas method in which each hierarchy includes spatial normalization using simpler pre-segmented structures in order to reduce the inter-patient variability of more complex target structures.
The proposed hierarchical method was evaluated with 19 muscles from 20 CT images of the hip and thigh using the manual segmentation by expert orthopedic surgeons as ground truth. The average symmetric surface distance was significantly reduced in the proposed method (1.53 mm) in comparison with the conventional method (2.65 mm).
We demonstrated that the proposed hierarchical multi-atlas method improved the accuracy of muscle segmentation from CT images, in which large inter-patient variability and insufficient contrast were involved.
KeywordsMulti-atlas label fusion Musculoskeletal segmentation Hierarchical strategy
This research was supported by MEXT/JSPS KAKENHI 26108004, 25242051 and 16K01411, JST PRESTO 20407, and AMED/ETH the strategic Japanese-Swiss cooperative research program.
Compliance with ethical standards
Conflicts of interest
The authors declare that they have no conflict of interest.
Human and animals rights
This study has been approved by the Institutional Review Board of Osaka University Hospital (No. 15056), where the patients’ medical data used in this work were obtained.
- 2.Arthofer C, Morgan PS, Pitiot A (2016) Hierarchical multi-atlas segmentation using label-specific embeddings, target-specific templates and patch refinement. In: Patch-based techniques in medical imaging LNCS 9993, pp 84–91Google Scholar
- 3.Baudin PY, Azzabou N, Carlier PG, Paragios N (2012) Prior knowledge, random walks and human skeletal muscle segmentation. In: International conference on medical image computing and computer-assisted intervention: MICCAI 2012, vol 15(Pt 1), pp 569–576Google Scholar
- 4.Diaz-Boladeras M, Angulo C, Domènech M, Albo-Canals J, Serrallonga N, Raya C, Barco A (2016) XIV Mediterranean conference on medical and biological engineering and computing 2016: MEDICON 2016. In: Conference on Medical and Biological Engineering and Computing, pp. 1179–1184Google Scholar
- 6.Fukuda N, Otake Y, Takao M, Yokota F, Ogawa T, Nakaya R, Tamura K, Grupp R, Farvardin A, Sugano N, Sato Y (2016) Statistical estimation of attachment of hip muscles based on measurement in cadavers. In: 16th annual meeting of CAOS-international proceedings, pp. 351–354Google Scholar
- 8.Glocker B, Komodakis N Drop. URL http://www.mrf-registration.net/deformable/index.html. Accessed 5 Apr 2018
- 13.Kamiya N, Muramatsu C, Zhou X, Chen H, Yokoyama R, Hara T (2013) Model-based approach to recognize the rectus abdominis muscle in CT by use of a virtual image-unfolding technique. IEICE Trans Inf Syst E–96–D(4):2–3Google Scholar
- 20.Ou Y, Doshi J (2012) Multi-atlas segmentation of the prostate: a zooming process with robust registration and atlas selection. MICCAI Grand Chall Prostate MR Image Segmen 7:1–7Google Scholar
- 23.Rueckert D, Schnabel J IRTK. https://biomedia.doc.ic.ac.uk/software/irtk/. Accessed 5 Apr 2018
- 25.Styner M, Lee J, Chin B, Chin M (2008) 3D segmentation in the clinic: a grand challenge II: MS lesion segmentation. Midas J 1–6Google Scholar
- 26.Takao M, Ogawa T, Yokota F, Otake Y, Hamada H, Sakai T, Sato Y, Sugano N (2017) Pre-operative fatty degeneration of gluteus minimus predicts falls after tha. Bone Joint J 99(SUPP 6):39–39Google Scholar
- 31.Yokota F, Okada T, Takao M, Sugano N, Tada Y, Tomiyama N, Sato Y (2013) Automated CT segmentation of diseased hip using hierarchical and conditional statistical shape models. In: International conference on medical image computing and computer-assisted intervention: MICCAI 2013, pp 190–197Google Scholar
- 32.Yokota F, Takaya M, Okada T, Takao M, Sugano N, Tada Y, Tomiyama N, Sato Y (2012) Automated muscle segmentation from 3D CT data of the hip using hierarchical multi-atlas method. In: The 12th annual meeting of the international society for computer assisted orthopaedic surgery: CAOS 2012, vol 1, pp 16–18Google Scholar