Enhancement of m-FISH Images using Spectral Unmixing

Breast carcinoma is after skin carcinoma the second most common form of cancer among women. Multicolour fluorescent in-situ hybridisation (m-FISH) is a common method for staging breast carcinoma. The interpretation of m-FISH images is complicated due to two effects: (i) Spectral overlap in the emission spectra of fluorochrome marked DNA probes and (ii) tissue autofluorescence. In this paper hyper-spectral images of m-FISH preparations are used and spectral unmixing is applied to produce false colour images with higher contrast and better information content than standard RGB images. The spectral unmixing is realised by combinations of: Orthogonal Projection Analysis (OPA), Alterating Least Squares (ALS), Simple-to-use Interactive Self-Modeling Mixture Analysis (SIMPLISMA) and VARIMAX. These are applied on the data to reduce tissue autofluorescence and resolve the spectral overlap in the emission spectra. The results show that spectral unmixing methods reduce the intensity caused by tissue autofluorescence by up to 78% and enhance image contrast by algorithmically reducing the overlap of the emission spectra. Keywords—breast carcinoma, hyperspectral imaging, m-FISH, spectral unmixing

[1]  D M Barnes,et al.  Overexpression of the c-erbB-2 oncoprotein: why does this occur more frequently in ductal carcinoma in situ than in invasive mammary carcinoma and is this of prognostic significance? , 1992, European journal of cancer.

[2]  N. Hynes,et al.  The biology of erbB-2/neu/HER-2 and its role in cancer. , 1994, Biochimica et biophysica acta.

[3]  R. Lupu,et al.  Interaction between erbB-receptors and heregulin in breast cancer tumor progression and drug resistance. , 1995, Seminars in cancer biology.

[4]  B. Ljung,et al.  The HER2 (c-erbB-2) oncogene is frequently amplified in in situ carcinomas of the breast. , 1992, Oncogene.

[5]  W. Windig,et al.  Interactive self-modeling mixture analysis , 1991 .

[6]  T. Visakorpi,et al.  Improved technique for analysis of formalin-fixed, paraffin-embedded tumors by fluorescence in situ hybridization. , 1994, Cytometry.

[7]  D. Massart,et al.  Orthogonal projection approach applied to peak purity assessment. , 1996, Analytical chemistry.

[8]  Ioannis Pitas,et al.  Automated evaluation of her-2/neu status in breast tissue from fluorescent in situ hybridization images , 2005, IEEE Transactions on Image Processing.

[9]  P. Carter,et al.  Differential responses of human tumor cell lines to anti-p185HER2 monoclonal antibodies , 1993, Cancer Immunology, Immunotherapy.

[10]  A. Muñoz-Barrutia,et al.  Blind Spectral Unmixing of M-FISH Images by Non-negative Matrix Factorization , 2007, 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[11]  H. Kaiser.  The varimax criterion for analytic rotation in factor analysis , 1958 .

[12]  Chein-I Chang,et al.  Semi-Supervised Linear Spectral Unmixing Using a Hierarchical Bayesian Model for Hyperspectral Imagery , 2008, IEEE Transactions on Signal Processing.