Medical Radiology and Radiation Safety. 2016. Vol. 61. No. 6. P. 64-67


N.E. Kosykh1, S.Z. Savin1, T.P. Potapova2

The Use of Textural Analysis for Assessment of Differences Between Metastatic and Non-Metastatic Zones on Planar Bone Scintigrams

1. Computer Center of Far-Eastern Branch, Russian Academy of Sciences, Khabarovsk, Russia, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it. ; 2. Far-Eastern State Medical University, Khabarovsk, Russia


Purpose: Study of metastatic images’ textural characteristics on planar scintigrams of skeleton.

Material and methods: The study involves computer analysis program for automatic assessment of skeletal metastases which is based on image recognition principles and is capable of expert analysis. The program’s functionality includes skeleton image segmentation, calculation of textural, histogrammic and morphometric parameters, creation of learning sample, dividing segmented foci to pathological and non-pathological by means of classifying function based on support vector machine. The study is based on planar scintigraphy data assessment of 168 patients with disseminated breast cancer. Computer automated analysis was used to distinguish pathological (metastatic) from physiological (non-metastatic) radiopharmaceutical hyperfixation foci in which Haralick’s textural features were determined: autocorrelation, contrast, forth feature and heterogeneity. On anterior view scintigrams, in a segmented skeleton, 8 zones were selected: scull, accessory sinuses of the nose, spine, sternum, thorax, pelvis, large joints and long cortical bones. On posterior view scintigrams 6 zones were selected: scull, spine, thorax, large joints and long cortical bones. Weighted means were calculated for each zone. Derived values for pathological and non-pathological hyperfixation foci were paralleled with calculation of Students’ criteria.

Results: In most skeletal zones textural Haralick’s features prevail in pathological hyperfixation foci over similar values in physiological hyperfixation foci. Differences by all 5 features between pathological and physiological radiopharmaceutical hyperfixation foci on frontal scintigrams were found in sternal and pelvic zones. On posterior scintigrams they were found only in pelvic zones. Most commonly in pathological hyperfixation foci (both posterior and anterior scintigrams) prevalence of contrast was found in comparison with similar features of physiological hyperfixation foci.

Conclusion: Results of our research show possibility of Haralick’s textures application for differential diagnostics of metastatic and non-metastatic foci on planar bone scintigrams by means of computer automated analysis.

Key words: automated computer analysis, image recognition, planar scintigrams, hyperfixation zone, radiopharmaceutical, histograms, image brightness


  1. Kalender V. Komp'yuternaya tomografiya. Moscow: Tekhnosfera. 2006. 343 p. (In Russ.).
  2. Pratt W.K. Digital Image Processing (4th edition). John Wiley & Sons, Inc. 2007. 807 p.
  3. Gritsenko N.N., Ul'yanychev N.V. Programmnyi kompleks intellektual'noi obrabotki mediko-biologicheskikh dannykh. Informatika i sistemy upravleniya. 2010. Vol. 24. No. 2. P. 76-80. (In Russ.).
  4. Gonsalens R., Vuds R., Eddins S. Tsifrovaya obrabotka izobrazhenii v srede MATLAB. Moscow: Tekhnosfera. 2006. 616 p. (In Russ.).
  5. Myasnikov V.V., Popov S.B., Sergeev V.V., Chernov V.M. Raspoznavanie izobrazhenii. V sb.: «Metody komp'yuternoi obrabotki izobrazhenii». Pod red. V.A. Soifera. Moscow: Fizmatlit. 2003. P. 251-300. (In Russ.).
  6. Fisenko V.T., Fisenko T.Yu. Komp'yuternaya obrabotka i raspoznavanie izobrazhenii. Saint Petersburg: SPbGU ITMO. 2008. 192 p. (In Russ.).
  7. Yane B. Tsifrovaya obrabotka izobrazhenii. Moscow: Tekhnosfera. 2007. 584 p. (In Russ.).
  8. Haralick R.M., Shanmugam K., Dinstein I. Textural features of image classification. IEEE Transact. Systems. Man and Cybernetics. 1973. Vol. 6. Suppl. SMC-3. P. 610-621.
  9. Gaidel' A.V., Pervushkin S.S. Issledovanie teksturnykh priznakov dlya diagnostiki zabolevanii kostnoi tkani po rentgenovskim izobrazheniyam Komp'yuternaya optika. 2013. Vol. 37. No. 1. P. 42-48. (In Russ.).
  10. Materka A., Cichy P., Tuliszkiewicz J. Texture analysis of x-ray images for detection of changes in bone mass and structure. In: Texture Analysis in Machine Vision. Series in Machine Perception and Artificial Intelligence. Ser. 40. Ed. M.K. Pietikainen. World Scientific. 2000. P. 185-189.
  11. Karahaliou A.N., Boniatis I.S., Skiadopoulos S.G. et al. Breast cancer diagnosis: analyzing texture of tissue surrounding microcalcifications. IEEE Transactions on Information Technology in Biomedicine. 2008. Vol. 12. No. 3. P. 731-738.
  12. Lee N.J. Computer-Aided diagnostic system for digital mammograms. A thesis of master of science in systems science. Jackson State University, December, 2006. 25 p.
  13. Gletsos M., Mougiakakou S.G., Matsopoulos G.K. et al. Computer-aided diagnostic system to characterize CT focal liver lesions: design and optimization of a neural network classifier IEEE Transact. Inform. Biomed. 2003. Vol. 7. No. 3. P. 153-162.
  14. Pasha S.P., Ternovoi S.K. Radionuklidnaya diagnostika. Moscow: GEOTAR-media. 2008. 204 p. (In Russ.).
  15. Sadik M., Suurkula M., Höglund P. et al. Improved classifications of planar whole-body bone scans using a computer-assisted diagnosis system: a multicenter, multiple-reader, multiple-case study. J. Nucl. Med. 2009. Vol. 50. No. 3. P. 368-375.
  16. Kovalenko V.L., Kosykh N.E., Savin S.Z., Gostyushkin V.V. Metody povysheniya effektivnosti komp'yuternykh avtomatizirovannykh tekhnologii v zadachakh radionuklidnoi diagnostiki Vrach i informatsionnye tekhnologii. 2013. No. 6. P. 42-48. (In Russ.).

For citation: Kosykh NE, Savin SZ, Potapova TP. The Use of Textural Analysis for Assessment of Differences between Metastatic and Non-Metastatic Zones on Planar Bone Scintigrams. 2016;61(6):64-7. Russian.

PDF (RUS) Full-text article (in Russian)