Abstract
We describe a decision support system to distinguish among hematology cases directly from microscopic specimens. The system uses an image database containing digitized specimens from normal and four different hematologic malignancies. Initially, the nuclei and cytoplasmic components of the specimens are segmented using a robust color gradient vector flow active contour model. Using a few cell images from each class, the basic texture elements (textons) for the nuclei and cytoplasm are learned, and the cells are represented through texton histograms. We propose to use support vector machines on the texton histogram based cell representation and achieve major improvement over the commonly used classification methods in texture research. Experiments with 3,691 cell images from 105 patients which originated from four different hospitals indicate more than 84% classification performance for individual cells and 89% for case based classification for the five class problem.
Similar content being viewed by others
References
Aisenberg A (1995) Coherent view of non-Hodgkin’s lymphoma. J Clin Oncol 13:2656–2675
Chabat F, Yang G, Hansell D (2003) Obstructive lung diseases: texture classification for differentiation at CT. Radiology 228:871–877
Chan J, Banks P, Cleary M, Delsol G, De Wolf-Peeters C, Falini B, Gatter K, Grogan T, Harris N, Isaacson P (1995) A revised European-American classification of lymphoid neoplasms proposed by the international lymphoma study group—a summary version. Am J Clin Pathol 103:543–560
Chen W, Meer P, Georgescu B, He W, Goodell LA, Foran D (2005) Image mining for investigative pathology using optimized feature extraction and data fusion. Comput Methods Programs Biomed 79:59–72
Comaniciu D, Meer P, Foran D (1999) Image-guided decision support system for pathology. Mach Vis Appl 11:213–224
Cortes C, Vapnik V (1995) Support vector networks. Mach Learn 20:273–297
Cotran R, Kumar V, Collins T, Robbins S (1994) Pathologic basis of disease, 5th edn. W.B. Saunders Company, Philadelphia
Cristianini N, Shawe-Taylor J (2000) Support vector machines and other kernel-based learning methods. Cambridge University Press, Cambridge
Cula O, Dana K (2004) 3D texture recognition using bidirectional feature histograms. Intl J Comput Vis 59(1)
Duda R, Hart P, Stork D (2001) Pattern classification, 2nd edn. Wiley, New York
Efros A, Leung T (1999) Texture synthesis by non-parametric sampling. In: Proceedings of 7th international conference on computer vision, Kerkyra, Greece, pp 1033–1038
Foran D, Comaniciu D, Meer P, Goodell L (2000) Computer-assisted discrimination among lymphomas and leukemia using imunophenotyping, intelligent image repositories and telemicroscopy. IEEE Trans Inform Tech Biomed 4:265–273
Friedman J, Hastie T, Tibshirani R (2000) Additive logistic regression: a statistical view of boosting. Ann Stat 28(2):337–407
Galloway R (1975) Texture analysis using gray level run lengths. Comput Graphic Image Process 4:172–179
Garcia-Conde J, Cabanillas F (1996) Cell lymphoma: a lymphoproliferatie disorder associated with aberrant function of the cell cycle. Leukemia 10:78–83
Haralick R (1979) Statistical and structural approaches to texture. IEEE 67:786–804
Haralick R, Shanmugam K, Dinstein I (1973) Textural features for image classification. IEEE Trans Syst Man Cybern 3:610–621
Hassan K, Tweed T, Miguet S (2005) A multi-resolution approach for content-based image retrieval on the grid-application to breast cancer detection. Methods Inform Med 44:211–214
Hastie T, Tibshirani R, Freidman J (2001) The elements of statistical learning. Springer, Heidelberg
Heeger D, Bergen J (1995) Pyramid-based texture analysis/synthesis. In: SIGGRAPH ’95: proceedings of the 22nd annual conference on computer graphics and interactive techniques, pp 229–238
Hertzmann A, Jacobs C, Oliver N, Curless B, Salesin D (2001) Image analogies. In: SIGGRAPH ’01: proceedings of the 28th annual conference on computer graphics and interactive techniques, pp 327–340
Hsu C, Lin C (2002) A comparison of methods for multiclass support vector machines. IEEE Trans Neural Netw 13:415–425
Huang Y, Chen J, SWC (2006) Diagnosis of hepatic tumors with texture analysis in nonenhanced computed tomography images. Acad Radiol 13:713–720
Ikeda T, Hagiwara M (2000) Content-based image retrieval system using neural networks. Int J Neural Syst 10:417–424
Jaulent M, Le Bozec C, Cao Y, Zapletal E, Degoulet P (2000) A property concept frame representation for flexible image-content retrieval in histopathology databases. In: AMIA symposium, pp 379–383
Joachims T (1999) Making large-scale SVM learning practical. advances in kernel methods—support vector learning. MIT Press, Cambridge
Kuhn F, Giardina C (1982) Elliptic Fourier features of a closed contour. Comput Graphics Image Process 18:236–258
Le Bozec C, Zapletal E, Jaulent M, HD, Degoulet P (2000) Towards content-based image retrieval in a his-integrated pacs. In: AMIA symposium, pp 477–481
Lehmann T, Wein B, Dahmen J, Bredno J, Vogelsang F, Kohnen M (2000) Content-based image retrieval in medical applications: a novel multistep approach. In: Proceedings of SPIE: storage and retrieval for media databases 2000, vol 3972, pp 312–320
Leung T, Malik J (1999) Recognizing surfaces using three-dimensional textons. In: Proceedings of 7th international conference on computer vision, Kerkyra, pp 1010–1017
Mao J, Jain A (1992) Texture classification and segmentation using multiresolution simultaneous autoregressive models. Pattern Recogn 25:173–188
Mavroforakis M, Georgiou H, Dimitropoulos N, Cavouras D, Theodoridis S (2006) Mammographic masses characterization based on localized texture and dataset fractal analysis using linear, neural and support vector machine classifiers. Artif Intell Med 37:145–162
Mudigonda N, Rangayyan R, Desautels J (2000) Gradient and texture analysis for the classification of mammographic masses. IEEE Trans Med Imaging 19:1032–1043
Nielsen B, Albregtsen F, Danielsen H (2004) Low dimensionality adaptive texture feature vectors from class distance and class difference matrices. IEEE Trans Med Imaging 23:73–84
Nielsen B, Danielsen H (2006) Prognostic value of adaptive textural features—the effect of standardizing nuclear first-order gray level statistics and mixing information from nuclei having different area. Cellular Oncol 28:85–95
Rajesh L, Pattari S, GG, Dey P, Srinivasan R (2004) Image morphometry of acute leukemias. comparison between lymphoid and myeloid subtypes. Anal Quant Cytol Histol 26:57–60
Ricketts I, Banda-Gamboa H, Cairns A, Hussein K (1992) Automatic classification of cervical cells-using the frequency domain. In: IEE coll. on app. of im. proc. in mass health screening, vol 9, pp 1–4
Rozman C, Montserrat E (1995) Chronic lymphocytic leukemia. New Engl J Med 333:1052–1057
Schmid C (2001) Constructing models for content-based image retreival. In: Proceedings of IEEE conference on computer vision and pattern recognition, Kauai, pp 39–45
Schmid K, Angerstein N, Geleff S, Gschwendtner A (2006) Quantitative nuclear texture features analysis confirms who classification 2004 for lung carcinomas. Modern Pathol 19:453–459
Schnorrenberg F, Pattichis C, Schizas C, Kyriacou K (2000) Content-based retrieval of breast cancer biopsy slides. Technol Health Care 8:291–297
Sheshadri H, Kandaswamy A (2007) Experimental investigation on breast tissue classification based on statistical feature extraction of mammograms. Comput Med Imaging Graph 31:46–48
Vadlamudi G (1996) Leukemic phase of mantle cell lymphoma: two case reports and review of the literature. Arch Pathol Lab Med 120:35–40
Varma M, Zisserman A (2002) Statistical approaches to material classification. In: Proceedings of 7th European conference on computer vision, Copehagen, Denmark
Varma M, Zisserman A (2003) Texture classification: are filter banks necessary?. In: Proceedings of 9th international conference on computer vision, Nice, France, vol 2, pp 691–698
Walker R, Jackway P (1996) Statistical geometric features-extensions for cytological texture analysis. In: Proceedings of 13th international conference on pattern recognition, vol 2, pp 790–794
Wyszecki G, Stiles W (1982) Concepts and methods, quantitative data and formulae. Wiley, New York
Xu C, Prince J (1998) Snakes, shapes, and gradient vector flow. IEEE Trans Image Process 7(3):359–369
Yang L, Meer P, Foran D (2005) Unsupervised segmentation based on robust estimation and color active contour models. IEEE Trans Inform Technol Biomed 9:475–486
Acknowledgments
This research was funded, in part, by grants from the NIH through contracts 5R01LM007455-03 from the National Library of Medicine and 5R01EB003587-02 from the National Institute of Biomedical Imaging and Bioengineering.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Tuzel, O., Yang, L., Meer, P. et al. Classification of hematologic malignancies using texton signatures. Pattern Anal Applic 10, 277–290 (2007). https://doi.org/10.1007/s10044-007-0066-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10044-007-0066-x