Ensemble of classifiers and wavelet transformation for improved recognition of Fuhrman grading in clear-cell renal carcinoma

• Autorzy: Kruk M. , Kurek J. , Osowski S. , Koktysz R. , Swiderski B. , Markiewicz T.

Identyfikatory

DOI

10.1016/j.bbe.2017.04.005

• Języki publikacji: EN

Abstrakty

EN

The paper presents an improved system to recognition of Fuhrman grading in clear-cell renal carcinoma using an ensemble of classifiers. The novelty of solution includes the segmentation applying wavelet transformation in preprocessing stage, application of few selection methods for feature generation and using the ensemble of classifiers in final recognition step. The wavelet transformation is a very efficient tool for image de-noising and enhancing the edges of cell nuclei. The important distinction to other approaches is that diagnostic features of nuclei, based on the texture, geometry, color and histogram, are selected by using few methods, each relying on different mechanism of selection. These different sets of features have enabled creating the ensemble of classifiers based on the support vector machine and random forest, both cooperating with them. Such approach has led to the significant increase of the quality factors in comparison to the best existing results: sensitivity (the average of this solution 94.3% compared to 91.5%) and specificity (the average 98.6% compared to 97.5%.

Słowa kluczowe

EN

ensemble of classifiers; feature selection; Fuhrman grading; wavelet transformation; SVM; random forest

PL

zespół klasyfikatorów; selekcja cech; skala Fuhrmana; transformacja falkowa; SVM; las losowy

• Wydawca: Nałęcz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences ; Elsevier
• Czasopismo: Biocybernetics and Biomedical Engineering
• Rocznik: 2017
• Tom: Vol. 37, no. 3
• Strony: 357--364
• Opis fizyczny: Bibliogr. 33 poz., rys., tab., wykr.

Twórcy

autor

Kruk M.

• Warsaw University of Life Sciences, Warsaw, Poland

autor

Kurek J.

• Warsaw University of Life Sciences, Warsaw, Poland

autor

Osowski S.

• Warsaw University of Technology, 00-661 Warsaw, Koszykowa 75, Poland; Military University of Technology, Warsaw, Poland

autor

Koktysz R.

• Military Institute of Medicine, Warsaw, Poland

autor

Swiderski B.

• Warsaw University of Life Sciences, Warsaw, Poland

autor

Markiewicz T.

• Warsaw University of Technology, Warsaw, Poland; Military Institute of Medicine, Warsaw, Poland

Bibliografia

• [1] Kontak JA, Campbell SC. Prognostic factors in renal cell carcinoma. Med Biol Eng Comput 2003;30:467–80.
• [2] Bostwick DG, Cheng L. Urologic surgical pathology. Philadelphia: Mosby Elsevier; 2008.
• [3] Perroud B, Ishimaru T, Borowsky AD, Weiss RH. Grade-dependent proteomics characterization of kidney cancer. Mol Cell Proteomics 2009;8:971–85.
• [4] Delahunt B. Advances and controversies in grading and staging of renal cell carcinoma. Modern Pathol 2009;22:24–36.
• [5] Fuhrman SA, Lasky LC, Limas C. Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol 1982;6:655–63.
• [6] Yeh FC, Parwani AV, Pantanowitz L, Ho C. Automated grading of renal cell carcinoma using whole slide imaging. J Pathol Inform 2014;5:23.
• [7] Champion A, Lu G, Walker M, Kothari S, Osunkoya AO, Wang MD. Semantic interpretation of robust imaging features for Fuhrman grading of renal carcinoma. Proc. Conf. IEEE Eng Med. Biol. Soc.. 2014. pp. 6446–9.
• [8] Chrom P, Stec R, Semeniuk-Wojtas A, Bodnar L, Spencer NJ, Szczylik C. Fuhrman grade and neutrophil-to-lymphocyte ratio influence on survival in patients with metastatic renal cell carcinoma treated with first-line tyrosine kinase inhibitors. Clin Genitourinary Cancer 2016;14:457–64.
• [9] Kruk M, Osowski S, Markiewicz T, Słodkowska J, Koktysz R, Kozłowski W, et al. Computer approach to recognition of Fuhrman grade of cells in clear-cell renal cell carcinoma. Analyt Quant Cytol Histol 2014;36(3):147–60.
• [10] Kruk M, Kurek J, Osowski S, Koktysz R. Improved computer recognition of Fuhrman grading system in analysis of clear-cell renal carcinoma. Proc. VIPIMAGE Conf., Canary Islands. 2015. pp. 221–6 (printed CRC Press/Balkema).
• [11] Huang PW, Lai YH. Effective segmentation classification for HCC biopsy images. J Pattern Recognition 2010;43:1550–63.
• [12] Kruk M, Osowski S, Koktysz R. Recognition and classification of colon cells applying the ensemble of classifiers. Comput Biol Med 2009;39:156–65.
• [13] Markiewicz T, Korzynska A, Kowalski A, Swiderska-Chadaj Z, Murawski P, Grala B, et al. MIAP – Web-based platform for the computer analysis of microscopic images to support the pathological diagnosis. Biocybernet Biomed Eng 2016;36 (4):597–609.
• [14] Nanni L, Brahnam S, Ghidoni S, Lumini A. Toward a general-purpose heterogenous ensemble for pattern classification. Comput Intell Neurosci 2015. http://dx.doi.org/10.1155/2015/909123.
• [15] Matlab user manual – Image processing toolbox. Natick: MathWorks; 2015.
• [16] Soille P. Morphological Image Analysis, Principles applications. Berlin: Springer; 2003.
• [17] Tan PN, Steinbach M, Kumar V. Introduction to data mining. Boston: Pearson Education Inc.; 2006.
• [18] Mubarak DM, Sathik MM, Beevi SZ, Revathy K. A hybrid region growing algorithm for medical image segmentation. Int J Comp Sci Inf Technol 2012;4(3):61–70.
• [19] Luc V, Soille P. Watersheds in digital spaces: an efficient algorithm based on immersion simulation. IEEE Trans Pattern Anal Mach Intell 1991;13(6):583–98.
• [20] Daubechies I. Ten lectures on wavelets. Philadelphia: SIAM; 1992.
• [21] Gonzalez RC, Woods RE. Digital Image Processing. second ed. Prentice Hall; 2002.
• [22] Jung CR, Scharcanski J. Robust watershed segmentation using wavelets. Image Vision Comput 2005;23:661–9.
• [23] Wagner T. Texture analysis. In: Jahne B, Haussecker H, Geisser P, editors. Book of Computer Vision Application. Boston: Academic Press; 1999 [chapter 10].
• [24] Kim TY, Coi HJ, Cha SJ, Choi HK. Study on texture analysis of renal cell carcinoma nuclei based on the Fuhrman grading system. Proc. IEEE Workshop Enterprise Networking Computing in Healthcare Industry; 2005.
• [25] Gianazza E, Chinello C, Mainini V, Cazzaniga M, Squeo V, Albo G, et al. Alterations of the serum peptidome in renal cell carcinoma discriminating benign malignant kidney tumors. J Proteomics 2012;76:125–40.
• [26] Guyon I, Elisseeff A. An introduction to variable and feature selection. J Mach Learn Res 2003;3:1157–82.
• [27] Hall M. Correlation-based feature selection for discrete and numeric class machine learning. Proc. 17th Intern. Conf. Machine Learning. San Francisco: Morgan Kaufmann Publishers; 2000. p. 359–66.
• [28] Liu H, Yu L. Feature selection for high-dimensional data: A fast correlation-based based filter solution. Proc. 20th Intern. Conf. Machine Leaning (ICML-03); 2003. pp. 856–63.
• [29] Michalewicz Z. Genetic Algorithms + Data Structures = Evolution Programs. Berlin: Springer; 1996.
• [30] Breiman L. Random forests. Machine Learning 2001;45:5–32.
• [31] Kuncheva L. Combining pattern classifiers: methods and algorithms. New York: Wiley; 2004.
• [32] Daliri MR. Combining extreme learning machines using support vector machines for breast tissue classification. Comput Methods Biomech Biomed Eng 2015;18(2):185–91.
• [33] Scholkopf B, Smola A. Learning with Kernels. Cambridge: MIT Press; 2002.

Uwagi

PL

Opracowanie ze środków MNiSW w ramach umowy 812/P-DUN/2016 na działalność upowszechniającą naukę (zadania 2017).