Mizuho Nishio^1,2*, Atsushi K. Kono³, Kazuhiro Kubo⁴, Hisanobu Koyama³, Tatsuya Nishii³, Kazuro Sugimura³

Show more
¹ Clinical PET Center, Institute of Biomedical Research and Innovation, Kobe, Japan.
² Division of Functional and Diagnostic Imaging Research, Department of Radiology, Kobe University Graduate School of Medicine, Kobe, Japan.
³ Division of Radiology, Department of Radiology, Kobe University Graduate School of Medicine, Kobe, Japan.
⁴ Center for Radiology and Radiation Oncology, Kobe University Hospital, Kobe, Japan.
Show less

Abstract: The purpose of this study was to develop methodology to segment tumors on 18F-fluorodeoxyg- lucose (FDG) positron emission tomography (PET) images. Sixty-four metastatic bone tumors were included. Graph cut was used for tumor segmentation, with segmentation energy divided into unary and pairwise terms. Locally connected conditional random fields (LCRF) were proposed for the pairwise term. In LCRF, three-dimensional cubic window with length L was set for each voxel, and voxels within the window were considered for the pairwise term. Three other types of segmentation were applied: region-growing based on 35%, 40%, and 45% of the tumor maximum standardized uptake value (RG35, RG40, and RG45, respectively), SLIC superpixels (SS), and region-based active contour models (AC). To validate the tumor segmentation accuracy, dice similarity coefficients (DSC) were calculated between the result of each technique and manual segmentation. Differences in DSC were tested using the Wilcoxon signed-rank test. Mean DSCs for LCRF at L = 3, 5, 7, and 9 were 0.784, 0.801, 0.809, and 0.812, respectively. Mean DSCs for the other techniques were: RG35, 0.633; RG40, 0.675; RG45, 0.689; SS, 0.709; and AC, 0.758. The DSC differences between LCRF and other techniques were statistically significant (p < 0.05). Tumor segmentation was reliably performed with LCRF.

Keywords: Image Segmentation, FDG-PET, Graph Cut

Cite this paper: Nishio, M. , Kono, A. , Kubo, K. , Koyama, H. , Nishii, T. , Sugimura, K. (2015) Tumor Segmentation on ¹⁸F FDG-PET Images Using Graph Cut and Local Spatial Information. Open Journal of Medical Imaging, 5, 174-181. doi: 10.4236/ojmi.2015.53022.

References

[1]   Foster, B., Bagci, U., Mansoor, A., Xu, Z. and Mollura, D.J. (2014) A Review on Segmentation of Positron Emission Tomography Images. Computers in Biology and Medicine, 50, 76-96.
http://dx.doi.org/10.1016/j.compbiomed.2014.04.014

[2]   Dibble, E.H., Alvarez, A.C., Truong, M.T., Mercier, G., Cook, E.F. and Subramaniam, R.M. (2012) 18F-FDG Metabolic Tumor Volume and Total Glycolytic Activity of Oral Cavity and Oropharyngeal Squamous Cell Cancer: Adding Value to Clinical Staging. Journal of Nuclear Medicine, 53, 709-715. http://dx.doi.org/10.2967/jnumed.111.099531

[3]   Paidpally, V., Chirindel, A., Chung, C.H., Richmon, J., Koch, W., Quon, H., et al. (2014) FDG Volumetric Parameters and Survival Outcomes after Definitive Chemoradiotherapy in Patients with Recurrent Head and Neck Squamous Cell Carcinoma. American Journal of Roentgenology, 203, W139-W145.
http://dx.doi.org/10.2214/ajr.13.11654

[4]   Alluri, K.C., Tahari, A.K., Wahl, R.L., Koch, W., Chung, C.H. and Subramaniam, R.M. (2014) Prognostic Value of FDG PET Metabolic Tumor Volume in Human Papillomavirus-Positive Stage III and IV Oropharyngeal Squamous Cell Carcinoma. American Journal of Roentgenology, 203, 897-903.
http://dx.doi.org/10.2214/AJR.14.12497

[5]   Zaidi, H., Vees, H. and Wissmeyer, M. (2009) Molecular PET/CT Imaging-Guided Radiation Therapy Treatment Planning. Academic Radiology, 16, 1108-1133.
http://dx.doi.org/10.1016/j.acra.2009.02.014

[6]   Zaidi, H. and El Naqa, I. (2010) PET-Guided Delineation of Radiation Therapy Treatment Volumes: A Survey of Image Segmentation Techniques. European Journal of Nuclear Medicine and Molecular Imaging, 37, 2165-2187.
http://dx.doi.org/10.1007/s00259-010-1423-3

[7]   Boykov, Y. and Funka-Lea, G. (2006) Graph Cuts and Efficient N-D Image Segmentation. International Journal of Computer Vision, 70, 109-131.
http://dx.doi.org/10.1007/s11263-006-7934-5

[8]   Zhang, K., Zhang, L., Song, H. and Zhou, W. (2010) Active Contours with Selective Local or Global Segmentation: A New Formulation and Level Set Method. Image and Vision Computing, 28, 668–676.
http://dx.doi.org/10.1016/j.imavis.2009.10.009

[9]   Achanta, R., Shaji, A., Smith, K., Lucchi, A., Fua, P. and Süsstrunk, S. (2010) SLIC Superpixels. EPFL Technical Report 149300.

[10]   Nishio M. Kono, A.K., Koyama, H., Nishii, T. and Sugimura, K. (2015) Tumor Segmentation on FDG-PET: Usefulness of Locally Connected Conditional Random Fields. SPIE Proceedings, 9413, 6 p.

[11]   Massoptier, L. and Casciaro, S. (2007) Fully Automatic Liver Segmentation through Graph-Cut Technique. Proceedings of 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Lyon, 22-26 August 2007, 5243-5246.
http://dx.doi.org/10.1109/iembs.2007.4353524

[12]   Hernando, D., Kellman, P., Haldar, J.P. and Liang, Z.P. (2010) Robust Water/Fat Separation in the Presence of Large Field Inhomogeneities Using a Graph Cut Algorithm. Magnetic Resonance in Medicine, 63, 79-90.

[13]   Chen, S.F., Cao, L.L., Liu, J.Z. and Tang, X.O. (2006) Automatic Segmentation of Lung Fields from Radiographic Images of SARS Patients Using a New Graph Cuts Algorithm. Proceedings of 18th International Conference on Pattern Recognition (ICPR), Hong Kong, 20-24 August 2006, 271-274.
http://dx.doi.org/10.1109/ICPR.2006.304

[14]   Ballangan, C., Wang, X., Fulham, M., Eberl, S. and Feng, D.D. (2013) Lung Tumor Segmentation in PET Images Using Graph Cuts. Computer Methods and Programs in Biomedicine, 109, 260-268.
http://dx.doi.org/10.1016/j.cmpb.2012.10.009

[15]   Han, D., Bayouth, J., Song, Q., Taurani, A., Sonka, M., Buatti, J., et al. (2011) Globally Optimal Tumor Segmentation in PET-CT Images: A Graph-Based Co-Segmentation Method. In: Székely, G. and Hahn, H.K., Eds., Information Processing in Medical Imaging, Chap. 22, Springer, Berlin, 245-256.
http://dx.doi.org/10.1007/978-3-642-22092-0_21

[16]   Otsu, N. (1979) A Threshold Selection Method from Gray-Level Histograms. IEEE Transactions on Systems, Man and Cybernetics, 9, 62-66.
http://dx.doi.org/10.1109/TSMC.1979.4310076

[17]   Lee, J.A. (2010) Segmentation of Positron Emission Tomography Images: Some Recommendations for Target Delineation in Radiation Oncology. Radiotherapy and Oncology, 96, 302-307.
http://dx.doi.org/10.1016/j.radonc.2010.07.003

[18]   Torigian, D.A., Lopez, R.F., Alapati, S., Bodapati, G., Hofheinz, F., van den Hoff, J., et al. (2011) Feasibility and Performance of Novel Software to Quantify Metabolically Active Volumes and 3D Partial Volume Corrected SUV and Metabolic Volumetric Products of Spinal Bone Marrow Metastases on 18F-FDG-PET/CT. Hellenic Society of Nuclear Medicine, 14, 8-14.

[19]   Kirov, A.S. and Fanchon, L.M. (2014) Pathology-Validated PET Image Data Sets and Their Role in PET Segmentation. Clinical and Translational Imaging, 2, 253-267.
http://dx.doi.org/10.1007/s40336-014-0068-9

[20]   Sridhar, P., Mercier, G., Tan, J., Truong, M.T., Daly, B. and Subramaniam, R.M. (2014) FDG PET Metabolic Tumor Volume Segmentation and Pathologic Volume of Primary Human Solid Tumors. American Journal of Roentgenology, 202, 1114-1119.
http://dx.doi.org/10.2214/AJR.13.11456