Abstract
In this paper, co-occurrence matrix based texture features are used to classify masses as benign or malignant. As (digitized) mammograms have high depth resolution (4096 gray levels in this study) and the size of a co-occurrence matrix depends on Q, the number of gray levels used for image intensity (depth) quantization, computation using co-occurrence matrices derived from mammograms can be expensive. Re-quantization using a lower value of Q is routinely performed but the effect of such procedure has not been sufficiently investigated. This paper investigates the effect of re-quantization using different Q. Four feature pools are formed with features measured on co-occurrence matrices with Q ∈{400}, Q ∈{100}, Q ∈{50} and Q ∈{400, 100, 50}. Classification results are obtained from each pool separately with the use of a genetic algorithm and the Fisher’s linear discriminant classifier. For Q ∈{400, 100, 50}, the best feature subsets selected by the genetic algorithm and of size k=6,7,8 have a leave-one-out area under the receiver operating characteristic (ROC) curve of 0.92, 0.93 and 0.94, respectively. Pairwise comparisons of the area index show that the differences in classification results for Q ∈{400, 100, 50} and Q ∈{50} are significant (p<0.06) for all k while that for Q ∈{400, 100, 50} and Q ∈{400} or Q ∈{100} are not significant.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Chan, H.-P., et al.: Phys. Med. Biol. 40, 857–876 (1995)
Gupta, S., Markey, M.K.: Med. Phys. 32(6), 1598–1606 (2005)
Hanley, J.A., McNeil, B.J.: Radiology. 143(1), 29–36 (1982)
Hanley, J.A., McNeil, B.J.: Radiology. 148(3), 839–843 (1983)
Haralick, R.M., et al.: IEEE Trans. Sys., Man, Cyb. SMC-3(6), 610–621 (1973)
Holland, J.: Adaptation in natural and artificial systems. MIT Press, Cambridge (1975)
Lee, G.N., Bottema, M.J.: Proc. 5th IWDM, pp. 259–263 (2001)
Lee, G.N., Bottema, M.J.: Proc. WDIM, APRS, pp. 105–109 (2003)
Metz, C.: Sem. Nucl. Med. 8, 283–298 (1978)
Mudigonda, N.R., et al.: IEEE Trans. on Med. Img. 19(10), 1032–1043 (2000)
Sahiner, S., et al.: Med. Phys. 25(4), 516–526 (1998)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2006 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Lee, G.N., Hara, T., Fujita, H. (2006). Classifying Masses as Benign or Malignant Based on Co-occurrence Matrix Textures: A Comparison Study of Different Gray Level Quantizations. In: Astley, S.M., Brady, M., Rose, C., Zwiggelaar, R. (eds) Digital Mammography. IWDM 2006. Lecture Notes in Computer Science, vol 4046. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11783237_45
Download citation
DOI: https://doi.org/10.1007/11783237_45
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-35625-7
Online ISBN: 978-3-540-35627-1
eBook Packages: Computer ScienceComputer Science (R0)