An automatic segmentation method for multispectral microscopic cervical cell images

We have been developing a computer-aided diagnosis (CAD) system for automatically recognizing cervical cancer cells from Papanicolaou smear. Considering that pathological changes of cervix can be indicated by the abnormity of the nucleus of intermediate cell, the key task of this system is to find the intermediate cells and segment the nucleus precisely. This paper presents a novel approach for automatic segmentation of microscopic cervical cell images using multispectral imaging techniques. In order to capture images at different wavelengths, a Liquid Crystal Tunable Filter (LCTF) device is used to provide wavelength selection from 400nm to 720nm with an increment of 10nm. Considering the spectral variances of background, nucleus and cytoplasm, background is extracted firstly from the microscopic images by calculating pixel intensity variance at 470nm, 530nm, 570nm, 580nm and 650nm. Then superficial cells are extracted apart from intermediate cells easily at 530nm 650nm because of the different pixel intensity distribution of the two kinds of cells at these two wavelengths. To segment the nucleus from intermediate cells, we adopt two procedures. Firstly, the nuclei are roughly segmented apart by using an iterative maximum deviation between-cluster algorithm. Secondly, a novel rigorous algorithm based on active contour model is adopted to achieve more exact nuclei segmentation. Using the method proposed in this paper, we did experiments on over 300 cervical smears, and the results show that this method is more robust and precise.

[1]  Richard M. Levenson,et al.  Spectral imaging and microscopy , 2000 .

[2]  George E. Meyer,et al.  Monitoring Food Safety, Agriculture, and Plant Health , 2004 .

[3]  Kacem Chehdi,et al.  Automatic image segmentation system through iterative edge - region co-operation , 2002, Image Vis. Comput..

[4]  M. S. Moran,et al.  Remote Sensing for Crop Management , 2003 .

[5]  Demetri Terzopoulos,et al.  Snakes: Active contour models , 2004, International Journal of Computer Vision.

[6]  Hong Zheng,et al.  Novel multispectral imaging microscope with applications to biomedicine , 2005, SPIE BiOS.

[7]  Yan Zhang,et al.  Analysis of blood and bone marrow smears using multispectral imaging analysis techniques , 2005, SPIE Medical Imaging.

[8]  C. Yang,et al.  Airborne multispectral imagery for mapping variable growing conditions and yields of cotton, grain sorghum, and corn , 2001 .

[9]  Andrew H. Gee,et al.  Adaptive segmentation of ultrasound images , 1999, Image Vis. Comput..

[10]  Zhang Hongbo.  Design of a Novel Automated Slide-loader System for Optical Microscope , 2005 .

[11]  Dike detection using active contour model , 2003, IGARSS 2003. 2003 IEEE International Geoscience and Remote Sensing Symposium. Proceedings (IEEE Cat. No.03CH37477).

[12]  D. Maktav,et al.  Remote sensing of urban areas , 2005 .

[13]  C. Balas,et al.  A novel optical imaging method for the early detection, quantitative grading, and mapping of cancerous and precancerous lesions of cervix , 2001, IEEE Transactions on Biomedical Engineering.

[14]  Shunsuke Yokozeki,et al.  Active contour model based on mathematical morphology , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).

[15]  Jing Li Wang,et al.  Color image segmentation: advances and prospects , 2001, Pattern Recognit..