ArticlesNo Access

AUTOMATED DIAGNOSIS OF BRAIN TUMOURS ASTROCYTOMAS USING PROBABILISTIC NEURAL NETWORK CLUSTERING AND SUPPORT VECTOR MACHINES

DIMITRIS GLOTSOS

Department of Medical Physics, University of Patras, Rio-Patras, 26500, Greece

Search for more papers by this author

JUSSI TOHKA

Tampere University of Technology, Institute of Signal Processing, Tampere, Finland

University of California Los Angeles, Department of Neurology, Laboratory of NeuroImaging, USA

Search for more papers by this author

PANAGIOTA RAVAZOULA

University Hospital of Patras, Department of Pathology, Rio-Patras, 26500, Greece

Search for more papers by this author

DIONISIS CAVOURAS

Technological Institute of Athens, Department of Medical Instruments Technology, Aigaleo, Athens, 12210, Greece

Search for more papers by this author

, and

GEORGE NIKIFORIDIS

Department of Medical Physics, University of Patras, Rio-Patras, 26500, Greece

Search for more papers by this author

https://doi.org/10.1142/S0129065705000013Cited by:29 (Source: Crossref)

Abstract

A computer-aided diagnosis system was developed for assisting brain astrocytomas malignancy grading. Microscopy images from 140 astrocytic biopsies were digitized and cell nuclei were automatically segmented using a Probabilistic Neural Network pixel-based clustering algorithm. A decision tree classification scheme was constructed to discriminate low, intermediate and high-grade tumours by analyzing nuclear features extracted from segmented nuclei with a Support Vector Machine classifier. Nuclei were segmented with an average accuracy of 86.5%. Low, intermediate, and high-grade tumours were identified with 95%, 88.3%, and 91% accuracies respectively. The proposed algorithm could be used as a second opinion tool for the histopathologists.

Keywords:

References

L. M. DeAngelis, New England Journal of Medicine 344, 114 (2001), DOI: 10.1056/NEJM200101113440207. Crossref, Medline, Web of Science, Google Scholar
W. Shapiro and J. Shapiro, Oncology 12, 233 (1998). Medline, Web of Science, Google Scholar
World Health Organization , International histological classification of tumours: Histological typing of tumours of the central nervous system , 2nd edn. ( Springer-Verlag , Berlin , 1993 ) . Google Scholar
K. Herfarth, S. Gutwein and J. Debus, Seminars in Surgical Oncology 20, 13 (2001), DOI: 10.1002/ssu.1012. Crossref, Medline, Web of Science, Google Scholar
R. A. Praysonet al., J. Neurological Sciences 175, 33 (2000), DOI: 10.1016/S0022-510X(00)00274-4. Crossref, Medline, Web of Science, Google Scholar
W. Coonset al., Cancer 79, 1381 (1997). Crossref, Medline, Web of Science, Google Scholar
P. Sallinen, Quantitative pathology in astrocytic tumours, PhD dissertation, University of Tampere (1999) . Google Scholar
N. Reinhold and W. Schlote, Analytical and Quantitative Cytology and Histopathology 25, 12 (2003). Medline, Web of Science, Google Scholar
C. Decaesteckeret al., J. Neurosurgery 86, 532 (1997). Crossref, Medline, Web of Science, Google Scholar
M. Scarpelliet al., Analytical and quantitative Cytology and Histology 16, 351 (1994). Medline, Web of Science, Google Scholar
A. V. Wangenheimet al., Grading of Gliomas in stereotactic Biopsies with Neural Networks, Proc. 4th conf. Artificial Intelligence in Medicine10 (1993) pp. 486–488. Google Scholar
N. Belacel and M. Boulassel, Artificial Intelligence in Medicine 12, 201 (2001), DOI: 10.1016/S0933-3657(00)00086-5. Web of Science, Google Scholar
P. Spyridonoset al., Analytical and Quantitative Cytology and Histology 26, 317 (2002). Google Scholar
O. Faugeras and W. Pratt, IEEE Transactions on Pattern Analysis and Machine Intelligence 14, 323 (1980). Google Scholar
D. Glotsoset al., A new approach to robust clustering by density estimation in an autocorrelation derived feature space, Proc. 6th NORDIC Signal Processing Symposium (2004) pp. 296–299. Google Scholar
D. Specht, Neural Networks 3, 109 (1980), DOI: 10.1016/0893-6080(90)90049-Q. Crossref, Web of Science, Google Scholar
N. Piliouraset al., Pattern Recognition Letters 25, 249 (2004), DOI: 10.1016/j.patrec.2003.10.006. Crossref, Web of Science, Google Scholar
K. K. Yiu, M. W. Mak and S. Y. Kung, Int. J. Neural Systems 12, 381 (2002), DOI: 10.1142/S0129065702001278. Link, Google Scholar
R. M. Halarick, K. Shanmugam and I. Dinstein, IEEE Transactions on Systems, Man and Cybernetics 3, 610 (1973). Google Scholar
M. M. Galloway, Computer Graphics and Image Processing 4, 172 (1975). Crossref, Google Scholar
D. Glotsoset al., Analytical and Quantitative Cytology and Histology 26, 77 (2004). Medline, Web of Science, Google Scholar
V. Kechman, Learning and Soft Computing (MIT, USA, 2001) pp. 121–189. Google Scholar
S. Theodoridis and K. Koutroubas, Pattern recognition 342, (1999). Google Scholar
I. Kalatziset al., Medical Informatics and the Internet in Medicine 28, 221 (2003), DOI: 10.1080/14639230310001613449. Crossref, Medline, Web of Science, Google Scholar
A. Ifantis and S. Papadimitriou, Int. J. Neural Systems 13, 315 (2003), DOI: 10.1142/S0129065703001674. Link, Google Scholar
T. Mouroutis, S. Roberts and A. Bharath, Bioimagning 6, 79 (1998). Crossref, Google Scholar
K.-M. Lee and N. Street, A fast and robust approach for automated segmentation of breast cancer nuclei, Proc. Second IASTED Int. Conf. Computer Graphics and Imaging (1999) pp. 42–47. Google Scholar
S. J. McKenna, Automated analysis of papanicolaou smears, Ph.D. Dissertation, University of Dundee (1994) . Google Scholar
P. Bamford and B. Lovell, Signal Processing 71, 203 (1998), DOI: 10.1016/S0165-1684(98)00145-5. Crossref, Web of Science, Google Scholar
P. Ranefallet al., Analytical Cellular Pathology 15, 145 (1997). Crossref, Medline, Web of Science, Google Scholar
X. Zhuanget al., IEEE Trans. Image Processing 5, 1293 (1996), DOI: 10.1109/83.535841. Crossref, Medline, Web of Science, Google Scholar
E. Parzen, Ann. Inst. Stat. Math. 33, 1065 (1962). Crossref, Web of Science, Google Scholar
H. Frigui and R. Krishnapuram, IEEE Trans. Pattern Anal. Machine lntell. 21, 450 (1999), DOI: 10.1109/34.765656. Crossref, Web of Science, Google Scholar
P. W. Holland and R. E. Welsch, Communication Statistics — Theory and Methods 9, 813 (1977). Google Scholar
W. Zheng and G. G. Yen, Int. J. Neural Systems 12, 149 (2002), DOI: 10.1142/S0129065702001230. Medline, Google Scholar
H. Hamid Muhammed, Int. J. Neural Systems 12, 425 (2002), DOI: 10.1142/S0129065702001321. Link, Google Scholar
P. Sykacek, Outliers and Bayesian inference, Proc. NC 98 (1998) pp. 973–978. Google Scholar
S. Hawkinset al., Neural Networks 170 (2002). Web of Science, Google Scholar