Abstract
A non-parametric algorithm is described for the construction of a binary decision tree classifier. This tree is used to correlate textural features, computed from ultrasonographic prostate images, with the histopathology of the imaged tissue. The algorithm consists of two parts; growing and pruning. In the growing phase an optimal tree is grown, based on the concept of mutual information. After growing, the tree is pruned by an alternating interaction of two data sets. Moreover, the structure and performance of the constructed tree are compared to the results using a slightly modified corresponding growing and pruning algorithm. The modified algorithm provides better retrospective and prospective classification results than the original algorithm. The use of the tree for automated cancer detection in ultrasonographic prostate images results in retrospective and prospective accuracy of 77.9% and 72.3%, respectively. Using this tissue characterisation, a supporting tool is provided for the interpretation of transrectal ultrasonographic images.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Bertermann, H., Loch, T. andGouge, J. (1989): ‘Computer-gesteuerte Ultraschallbild-Analyse,’Urologie Nephrologie,1, pp. 24–30
Brawer, M. K. (1993): ‘The diagnosis of prostatic carcinoma,’Cancer,71, pp. 899–905
Breiman, L., Friedman, J. H., Olshen, R. A. andStone, C. J. (1984): ‘Classification and regression trees’ (Wadsworth International Group, Belmont, California, USA)
Chodak, G. W., Wald, V., Parmer, E., Watanabe, H., Ohe, H., andSaitoh, M. (1986): ‘Comparison of digital examination and transcrectal ultrasonography for the diagnosis of prostatic cancer,’J. Urol.,135, pp. 951–954
Friedman, J. H. (1977): ‘A recursive partitioning decision rule for nonprametric classification,’IEEE Trans.,C-26, pp. 404–408
Gelfand, S. B., Ravishankar, C. S. andDelp, E. J. (1991): ‘An iterative growing and pruning algorithm for classification tree design,’IEEE Trans.,PAMI-13, pp. 163–174
Gerber, G. S., Goldberg, R. andChodak, G. W. (1992): ‘Local staging of prostate cancer by tumor volume, prostate-specific antigen, and transrectal ultrasound,’Urol.,40, pp. 311–316
Giesen, R. J. B. andHuynen, A. L. (1991): ‘Ultrasonographic tissue discrimination; automatic detection of prostate carinoma’. Master Thesis, Department of Computer Science, University of Twente, Enschede, The Netherlands
Giesen, R. J. B., Huynen, A. L., de la Rosette, J. J. M. C. H., Schaafsman, H. E., van Lersel, M. P., Aarnink, R. G., Debruyne, F. M. J. andWijkstra, H. (1994): ‘The reliability of computer analysis of ultrasonographic prostate images; the influence of inconsistent histopathology,’Ultrasound Med. Biol.,20, pp. 871–876
Haralick, R. M., Shanmugam, K. andDinstein, I. (1973): ‘Textural features for image classification,’IEEE Trans.,SMC-3, pp. 610–621
Henrichon, E. G. andFu, K. S. (1969): ‘A nonparametric partitioning procedure for pattern classification,’IEEE Trans.,C-18, pp. 614–624
Huynen, A. L., Giesen, R. J. B., de la Rosetta, J. J. M. C. H., Aarnink, R. G., Debruyne, F. M. J. andWijkstra, H. (1994): ‘Analysis of ultrasonographic prostate images for the detection of prostate carcinoma: the Automated Urologic Diagnostic EXpert system,’Ultrasound Med. Biol.,20, pp. 1–10
Kaye, K. W. andLightner, D. J. (1993): ‘Prostate biopsy indices: toward efficient use of transrectal ultrasound,’Urol.,41, pp. 417–420
Landeweerd, G. H., Timmers, T. andGelsema, E. S. (1983): ‘Binary tree versus single level tree classification of white blood cells,’Patt Recognit.,16, pp. 571–577
Lee, F., Torp-Pedersen, S. T. andMcLeary, R. D. (1989a): ‘Diagnosis of prostate cancer by transrectal ultrasound,’Urol. Clin. North Am.,16, pp. 663–673
Lee, F., Torp-Pedersen, S. T., Siders, D. B., Littrup, P. J. andMcLeary, R. D. (1989b): ‘Transrectal ultrasound in the diagnosis and staging of prostatic carcinoma,’Radiol. 170, pp. 609–615
Lin, Y. K. andFu, K. S. (1983): ‘Automatic classification of cervical cells using a binary tree classifier,’Patt. Recognit.,16, pp. 69–80
Loch, T., Gettys, T., Cochran, J. S., Fulgham, P. F. andBertermann, H. (1990): ‘Computer-aided image analysis in transrectal ultrasound of the prostate,’World J. Urol.,8, pp. 150–153
Mui, J. K. andFu, K. S. (1993): ‘Automatic classification of nucleated blood cells using a binary tree classifier,’IEEE Trans.,PAMI-2, pp. 429–443
Park, Y. andSklansky, J. (1990): ‘Automated design of liner tree classifiers,’Patt. Recognit.,23, pp. 1393–1412
Rounds, E. M. (1980): ‘A combined nonparametric approach to feature selection and binary tree design,’Patt. Recognit.,12, pp. 313–317
Scardino, P. T., Shinohara, K., Wheeler, T. M. andCarter, S. St. C. (1989): ‘Staging of prostate cancer; value of ultrasonography,’Urol. Clin. North Am.,16, pp. 713–734
Sethi, I. K. andSarvarayudu, G. P. R. (1982): ‘Hierarchical classifier design using mutual information,’IEEE Trans.,PAMI-4, pp. 441–445
Shinohara, K., Scardino, P. T., Carter, S. St. C. andWheeler, T. M. (1989): ‘Pathologic basis of the sonographic appearance of the normal and malignant prostate,’Urol. Clin. North., Am.,16, pp. 675–691
Wang, Q. R. andSuen, C. Y. (1987): ‘Large tree classifier with heuristic search and global training,’IEEE Trans.,PAMI-9, pp. 91–102
Wolf, J. S., Shinohara, K. andNarayan, P. (1992): ‘Staging of prostate cancer. Accuracy of transrectal ultrasound enhanced by prostate-specific antigen,’Br. J. Uroll 70, pp. 534–541
Zielhe, Th., Nauth, P., Stein, N., Seelen, V.W. andLoch, E.-G. (1985): ‘Quantitative Verfahren bei der Ultraschalldiagnostik,’Radiologe,25, pp. 468–473
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Giesen, R.J.B., Huynen, A.L., Aarnink, R.G. et al. Construction and application of hierarchical decision tree for classification of ultrasonographic prostate images. Med. Biol. Eng. Comput. 34, 105–109 (1996). https://doi.org/10.1007/BF02520013
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF02520013