Segmentation of Cancer Cells in Microscopic Images using Neural Network and Mathematical Morphology

This paper presents the segmentation of cancer cells in a microscopic tissue image from breast cancer. We perform color classification using the neural network. Subsequently, morphological operations and cell size considerations are used for eliminating spike noise and separating cancer cells. The excellent segmentation results from the proposed algorithm are demonstrated with microscopic images under both low and high histological noise conditions. These preliminary results of the automated image analysis show a promising solution to the traditional manual analysis

[1]  Arthur C. Sanderson,et al.  A System for Automated Liver Tissue Image Analysis: Methods and Results , 1985, IEEE Transactions on Biomedical Engineering.

[2]  Pornchai Phukpattaranont,et al.  AUTOMATIC CLASSIFICATION OF CANCER CELLS IN MICROSCOPIC IMAGES , 2006 .

[3]  Constantine Katsinis,et al.  Automated identification of microstructures on histology slides , 2004, 2004 2nd IEEE International Symposium on Biomedical Imaging: Nano to Macro (IEEE Cat No. 04EX821).

[4]  W. Hsu,et al.  On the accurate counting of tumor cells , 2003, IEEE Transactions on NanoBioscience.

[5]  K. Terai.  A Novel Direct Digital Frequency Synthesis with High Frequency Accuracy and Spectral Purity , 2006, 2006 SICE-ICASE International Joint Conference.

[6]  Hyoung-Eui Kim,et al.  Practical Selection Criteria for Determination of Confidence Level in Substantiation Test , 2006, 2006 SICE-ICASE International Joint Conference.

[7]  Martin T. Hagan,et al.  Neural network design , 1995 .