透過微陣列實驗，研究者可以同時觀察成千上萬個基因表現值。由於基因表現量的正確與否影響後續的分析甚鉅，如何在實驗的影像處理過程中得到可靠的基因表現量，便成研究是否成功的關鍵。在本研究中我們提出一個快速且準確的基因表現估計值。此方法利用三個k-means分群法剔除可能的污點並且加入了額外的空間訊息來改善分類的結果。從模擬的結果顯示，所提出的方法不僅克服了傳統之k-means分群法可能遭遇的問題，而且可以處理其他方法無法得到正確估計值的情況。經由實際資料的分析，在以RT-PCR為真正基因表現值下，發現本文中所提出的方法在基因表現值估計上比現有的方法有更好的績效。本研究亦提供一個品質指標，我們可用它來衡量估計出來的基因表現值是否可靠。

Microarray experiments allow people to investigate thousands of genes simultaneously. To be able to derive a reliable ratio estimate, therefore, becomes an important issue for the researchers. Image analysis, which obtains quantitative gene expression data, is crucial in the ensuring analyses if valid results are to be guaranteed. In this study, an efficient and accurate ratio estimation based on k-means clustering methods is developed. We use a three-step method to remove the suspicious artifact, and the additional grid information is included to improve the efficiency of segmentation. Results from the simulated study show that the proposed method does not only overcome the disadvantage of k-means based method, but also show its superiority to other methods. Using real-time polymerase chain reaction (RT-PCR) data as the golden standard and sum of square error (SEE) as the comparison criteria, our method outperforms the existing methods. A useful quality metric to determine the acceptance of our ratio estimation is also provided. It helps to determine the quality of the spot.

Adams, R. and Bischof, L. (1994), “Seeded region growing”, IEEE Transactions on Pattern Analysis and Machine Intelligence, 16(6), 641–647.

Ahmed, A. A., Vias, M., Iyer, N. G., Caldas, C., and Brenton, J. D. (2004), Microarray segmentation methods significantly influence data precision”, Nucleic Acids Research, 32(5), e50.

Baek, J., Son, Y. S., and McLachlan, G. J. (2007), “Segmentation and intensity estimation of microarray images using a gamma-t mixture model”, Bioinformatics, 23(4), 458–465.

Bozinov, D. and Rahnenfuhrer, J. (2002), “Unsupervised technique for robust target separation and analysis of dna microarray spots through adaptive pixel clustering”, Bioinformatics, 18(5), 747–756.

Brown, C. S., Goodwin, P. C., and Sorger, P. K. (2001), “Image metrics in the statistical analysis of dna microarray data”, Proceedings of the National Academy of Sciences of the United States of America, 98(16), 8944–8949.

Chen, Y., Dougherty, E. R., and Bittner, M. L. (1997), “Ratio-based decisions and the quantitative analysis of cdna microarray images”, Journal of Biomedical Optics, 2(4), 364–374.

Eisen, M.B. (1999), “Scanalyze”, http://rana.lbl.gov/EisenSoftware.htm for software and documentation.

Ekstrøm, C. T., Bak, S., Kristensen, C., and Rudemo, M. (2004), “Spot shape modelling and data transformations for microarrays”, Bioinformatics, 20(14), 2270–2278.

Kosinski, C., Li, S. W., Chan, S. Y., Zhang, J., Ho, C., Tsui, W. Y., Chan, T. L., Mifflin, R. C., Powell, D. W., Yuen, S. T., Leung, S. Y., and Chen, X. (2007), “Gene expression patterns of human colon tops and basal crypts and bmp antagonists as intestinal stem cell niche factors”, PNAS, 104(39), 15418–15423.

Lehmussola, A., Ruusuvuori, P., and Yli-Harja, O. (2006), “Evaluating the performance of microarray segmentation algorithms”, Bioinformatics, 22(23), 2910–2916.

Li, Q., Fraley, C., Bumgarner, R. E., Yeung, K. Y., and Raftery, A. E. (2005),“Donuts, scratches and blanks: robust model-based segmentation of microarray images”, Bioinformatics, 21(12), 2875–2882.

Novikov, E. and Barillot, E. (2005), “A robust algorithm for ratio estimation in two-color microarray experiments”, Journal of Bioinformatics and Computational Biology, 3(6), 1411–1428.

Rahnenfuhrer, J. and Bozinov, D. (2004), “Hybrid clustering for microarray image analysis combining intensity and shape features”, BMC Bioinformatics, 5(47).

Yang, Y. H., Buckley, M. J., D., S., and Speed, T. P. (2002), “Comparison of methods for image analysis on cdna microarray data”, Journal of Computational and Graphical Statistics, 11(1), 1–29.