Feature representation and signal classification in fluorescence in-situ hybridization image analysis

被引:30
作者
Lerner, B [1 ]
Clocksin, WF
Dhanjal, S
Hultén, MA
Bishop, CM
机构
[1] Ben Gurion Univ Negev, Dept Elect & Comp Engn, IL-84105 Beer Sheva, Israel
[2] Univ Cambridge, Comp Lab, Cambridge CB2 3QG, England
[3] Univ Warwick, Dept Biol Sci, Coventry CV4 7AL, W Midlands, England
[4] Microsoft Res, Cambridge CB2 3NH, England
来源
IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART A-SYSTEMS AND HUMANS | 2001年 / 31卷 / 06期
基金
英国工程与自然科学研究理事会;
关键词
color image segmentation; feature representation; fluorescence in-situ hybridization; image analysis; neural networks; signal classification;
D O I
10.1109/3468.983421
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Fast and accurate analysis of fluorescence in-situ hybridization (FISH) images for signal counting will depend mainly upon two components: a classifier to discriminate between artifacts and valid signals of several fluorophores (colors), and well discriminating features to represent the signals. Our previous work has focused on the first component. To investigate the second component, we evaluate candidate feature sets by illustrating the probability density functions (pdfs) and scatter plots for the features. The analysis provides first insight into dependencies between features, indicates the relative importance of members of a feature set, and helps in identifying sources of potential classification errors. Class separability yielded by different feature subsets is evaluated using the accuracy of several neural network (NN)-based classification strategies, some of them hierarchical, as well as using a feature selection technique making use of a scatter criterion. The complete analysis recommends several intensity and hue features for representing FISH signals. Represented by these features, around 90% of valid signals and artifacts of two fluorophores are correctly classifled using the NN. Although applied to cytogenetics, the paper presents a comprehensive, unifying methodology of qualitative and quantitative evaluation of pattern feature representation essential for accurate image classification. This methodology is applicable to many other real-world pattern recognition problems.
引用
收藏
页码:655 / 665
页数:11
相关论文
共 15 条
[1]  
[Anonymous], KNOWLEDGE BASED INTE
[2]  
Bishop C. M., 1995, NEURAL NETWORKS PATT
[3]  
Carter N. P., 1996, Bioimaging, V4, P41, DOI 10.1002/1361-6374(199606)4:2<41::AID-BIO2>3.3.CO
[4]  
2-O
[5]  
Castleman K. R., 1995, Bioimaging, V3, P88, DOI 10.1002/1361-6374(199506)3:2<88::AID-BIO5>3.3.CO
[6]  
2-T
[7]  
Fukunaga K., 1990, INTRO STAT PATTERN R
[8]   Feature selection: Evaluation, application, and small sample performance [J].
Jain, A ;
Zongker, D .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1997, 19 (02) :153-158
[9]  
John G.H., 1995, Eleventh Conference on Uncertainty in Artificial Intelligence, San Mateo, DOI [10.1109/TGRS.2004.834800, DOI 10.1109/TGRS.2004.834800, 10.5555/2074158.2074196, DOI 10.5555/2074158.2074196]
[10]  
KLINGER K, 1992, AM J HUM GENET, V51, P55