This paper presents a multispectral microscopy system for quantitative cytology. The automated system aims to classify microscopic samples for the purpose of disease diagnosis. While conventional practices rely on the analysis of grey scale or RGB color images, presented system uses thirty one spectral bands for analysis. Algorithms designed to enable image acquisition, image segmentation, feature extraction, and classification are presented. Results are presented for the problem of discriminating among four cell types. In addition, classification performance is compared to the case where multispectral information is not taken into consideration. Results show that the developed system and the use of multispectral information along with morphometric information extracted from spectral images can significantly improve the classification performance and aid in the process of disease diagnosis.
References
1.
WuH.S.BarbaJ., and GilJ., “A parametric fitting algorithm for segmentation of cell images,”IEEE Trans Biomed Eng, vol. 45, pp. 400–407, 1998.
2.
KimE., “A high-resolution multi-spectral imaging system for small satellites,”Acta Astronautica, vol. 52(9–12), pp. 813–818, 2003.
3.
KerekesJ. and BaumJ., “Spectral imaging system analytical model for subpixel object detection,”IEEE Transactions on Geoscience and Remote Sensing, vol. 40(5), pp. 1088–1101, 2002.
4.
KulckeA., “On-line classification of synthetic polymers using near infrared spectral imaging,”Journal of near infrared spectroscopy, vol. 11(1), pp. 71–81, 2003.
5.
TsunodaS., “Novel spectral imaging system combined with confocal laser microscopy for fish analysis of interphase nuclei,”Cytometry, p. 139, 2002.
6.
MaitiD.SennouneS., and Martinez-ZaguilanR., “Proton gradients in human breast cancer cells determined by confocal, multiphoton, and spectral imaging microscopy,”FASEB Journal, vol. 17(4), pp. A467, 2003.
7.
HaraguchiT., “Spectral imaging fluorescence spectroscopy,”Genes to Cells, vol. 7(9), pp. 881–887, 2002.
8.
RoyK.BottrillI.IngramsD.PankratovM.RebeizE.WooP.ShapshayS.KabaniS., and ManoharanR., “Diagnostic fluorescence spectroscopy of oral mucosa,” in Proceedings of SPIE, 1995, vol. 2395, pp. 135–142.
9.
GariniY.KatzirN.CabibD.BuckwaldR.SoenksenD., and MalikZ., Fluorescence Imaging Spectroscopy and Microscopy, chapter Spectral Bio-Imaging, Chemical Analysis Series. John Wiley & Sons, New York, 1996.
10.
FentonB.RofstadE.DegnerF., and SutherlandR., “Cryospectrophotometric determination of tumor intravascular oxyhemoglobin saturations: Dependence on vascular geometry and tumor growth,”Journal of Natl. Cancer Inst., Vol. 80, pp. 1612–1619, 1988.
11.
HaraguchiT.DingD.YamamotoA.KanedaT.KoujinT., and HiraokaY., “Multiple-color fluorescence imaging of chromosomes and microtubules in living cells,”Cell Structure Funct., Vol. 24, pp. 291–298, 1999.
12.
HiraokaY.SwedlowJ.PaddyM.AgardD., and SedatJ., “3D multiple-wavelength fluorescence microscopy for the structural analysis of biological phenomena,”Semin Cell Biology, Vol. 2, pp. 153–165, 1991.
13.
SchrockE.Du ManoirS.VeldmanT.SchoellB.WeinbergJ.Ferguson-SmithM.NingY.LedbetterD.Bar-AmI.SoenksenD.SariniY., and ReidT., “Multicolor spectral karyotyping of human chromosomes,”Science, vol. 273, pp. 494–497, 1996.
14.
RothmannC.Bar-AmI., and MalikZ., “Spectral imaging for quantitative histology and cytogenetics,”Histol Histopathol, vol. 13, pp. 921–926, 1998.
15.
OrnbergR. L.WoernerM., and EdwardsD. A., “Analysis of stained objects in histopathological sections by spectral imaging and differential absorption,”J. Histochem. Cytochem., vol. 47, no. 10, pp. 1307–1313, 1999.
16.
RothmannC.CohenA., and MalikZ., “Chromatin condensation in erythropoiesis resolved by multipixel spectral imaging: Differentiation versus apoptosis,”Journal of Histochem. Cytochem., vol. 45(8), pp. 1097–1108, 1997.
17.
FarkasD.BallouB.FisherF., and FishmanD., “Microscopic and mesoscopic spectral bio-imaging,” in Proceedings of SPIE, 1996, vol. 2678, pp. 200–209.
18.
MalikZ.DishiM., and GariniY., “Fourier transform multipixel spectroscopy and spectral imaging of protoporphyrin insingle melanoma cells,”Photochem. Photobiol., vol. 63, pp. 608–614, 1996.
19.
MalikZ.CabibD.BuckwaldR.TalmiA.GariniY., and LipsonS., “Fourier transform multipixel spectroscopy for quantitative cytology,”Journal of microscopy, vol. 182, pp. 133–140, 1996.
20.
UlrichI. and NourbakhshI., “Appearance-based place recognition for topological localization,” in Proceeding of the IEEE International Conference on Robotics and Automation, 2000.
21.
SwainM. and BallardD., “Color indexing,”International Journal of Computer Vision, vol. 7(1), pp. 11–32, 1991.
22.
PressW.TeukolskyS.VetterlingW., and FlanneryB., Numerical Recipes in C, Cambridge University Press, Boston, 1992.
23.
SchieleB. and CrowleyJ., “Object recognition using multidimensional receptive field histograms,” in Proceeding of Fourth European Conference on Computer Vision, 1996, pp. 14–16.
24.
CastlemanK. R., Digital Image Processing, Prentice Hall, 1979.
25.
WuH. S.BerbaJ., and GilJ., “Iterative thresholding for segmentation of cells from noisy images,”Journal of Microscopy, vol. 179, no. 3, pp. 296–304, 2000.
26.
Meas-YedidVannaryTilieSorin, and Olivo-MarinJean-Christophe, “Color image segmentation based on markov random field clustering for histological image analysis,” in ICPR '02: Proceedings of the 16 th International Conference on Pattern Recognition (ICPR'02)Volume 1, Washington, DC, USA, 2002, p. 10796, IEEE Computer Society.
27.
SahooP. K.SoltaniS., and WongA. K. C., “A survey of thresholding techniques,”Computer Vision, Graphics and Image Processing, vol. 41, pp. 233–260, 1988.
28.
AdamsR. and BischofL., “Seeded region growing,”IEEE Trans. Pattern Analys. Machine Intell., vol. 16, pp. 641–647, June 1994.
29.
LieW.N., “Automatic target segmentation by locally adaptive image thresholding,”IEEE Trans. Image Process., vol. 4, pp. 1036–1041, July 1995.
30.
SerP.K. and SiuW.C., “Novel detection of conics using 2-D Hough planes,” in IEEE Proc. Vision, Image Signal Processing, 1995, vol. 142, pp. 262–270.
31.
AgamG. and DinsteinI., “Geometric separation of partially overlapping nonrigid objects applied to automatic chromosome classi-fication,”IEEE Trans. Pattern Anal. Mach. Intell., vol. 19, no. 11, pp. 1212–1222, November 1997.
32.
OsherS. and SethianJ., “Fronts propagating with curvature dependent speed: Algorithms based on Hamilton-Jacobi formulations,”Journal of Computationl Physics, pp. 12–49, 1988.
33.
BishopC. M., Neural Networks for Pattern Recognition, Oxford University Press, New York, 1995.
34.
ChooJ.JiamthapthaksinR.ChenC.CelepcikayO.GiustiC., and EickC., “Mosaic: A proximity graph approach to agglomerative clustering,” in Proc. 9th International Conference on Data Warehousing and Knowledge Discovery (DaWaK), 2007.
35.
ShahS. and AggarwalJ. K., “Multiple feature integration for robust object localization.,” in Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 1998, pp. 765–771.
36.
NairD. and AggarwalJ. K., “A focused target segmentation paradigm,” in Fourth European Conference on Computer Vision, Cambridge, UK, April 1996, vol. 1, pp. 579–588.
37.
GonzalesR. and WoodsR., Digital Image Processing, Addison-Wesley Publishing Company, 1992.
38.
KassM.WitkinA., and TerzopoulosD., “Snakes, active contour model,”International Journal of Computer Vision, pp. 321–331, 1988.
39.
VeseL. and ChanT., “A multiphase level set framework for image segmentation using the Mumford and Shah model,”International Journal of Computer Vision, no. 3, pp. 271–293, 2002.
40.
MorelJ. and SoliminiS., Progress in nonlinear differential equations and their applications, vol. 14, chapter Variational Methods in Image Segmentation, Birkhauser, Boston, 1995.
41.
RoussonM. and DericheR., “A variational framework for active and adaptive segmentation of vector valued images,” in Proc. of IEEE Workshop on Motion and Video Computing, 2003.
42.
MumfordD. and ShahJ., “Optimal approximation by piecewise smooth functions and associated variational problems,”Communication Pure and Applied Mathematics, p. 577, 1989.
43.
TangY.LiX.von FreybergA., and GochG., “Automatic segmentation of the papilla in a fundus image based on the c-v model and a shape restraint,” in Proc. of International Conference on Pattern Recognition, 2006, pp. I: 183–186.
44.
PuzichaJ.HofmannT., and BuhmannJ. M., “Histogram clustering for unsupervised segmentation and image retrieval,”Pattern Recognition Letters, vol. 20, no. 8, pp. 899–909, Aug. 1999.
45.
CastlemanK. R., Digital Image Processing, Prentice Hall, 1998.
46.
JohnG. H. and LangleyP., “Estimating continuous distributions in bayesian classifiers,” in Proceedings of the Eleventh Conference on Uncertainty in Artificial Intelligence, 1995, pp. 338–345.
47.
FriedmanN.GeigerD., and GoldszmidtM., “Bayesian network classifiers,”Machine Learning, vol. 29, pp. 131–163, 1997.
48.
MitchellT., Machine Learning, McGraw Hill, 1997.
49.
BreimanL., “Bagging predictors,”Machine Learning, vol. 24, pp. 123–140, 1996.
50.
DudaR. O.HartP. E., and StorkD. G., Pattern Classification, Wiley, 2001.
