Review on Image Segmentation Methods in cDNA Microarray Experiments and a Novel Algorithm for Segmentation
Kalesh M. Karun1, Binu V. S2, Kala M. Karun3, Keerthana Prasad4, Nair N. S5, K. Manjunatha Prasad6, K. M. Girisha7

1Kalesh M. Karun, PhD Scholar, Department of Statistics, Manipal University, Manipal, India.
2*Binu V. S, Assoc. Prof., Department of Statistics, Manipal University, India.
3Kala M. Karun, Asst. Prof., Department of Information and Technology, Adi Shankara Institute of Engineering and Technology, India.
4Keerthana Prasad, Assoc. Prof., Manipal Center for Information Science, Manipal University, Manipal, India.
5N. Sreekumaran Nair, Prof., and Head, Department of Statistics, Manipal University, Manipal, India.
6K. Manjunatha Prasad, Prof., Department of Statistics, Manipal University, Manipal, India.
7K. M. Girisha, Prof., & Head, Department of Medical Genetics, KMC Manipal, Manipal University, India.
Manuscript received on March 12, 2015. | Revised Manuscript received on March 13, 2015. | Manuscript published on March 25, 2015. | PP:19-24 | Volume-3 Issue-5, March 2015. | Retrieval Number: E0930033515

Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (

Abstract: Microarray experiments are used to measure gene expression levels of thousands of genes at a time. The image analysis has an important role in the microarray data analysis and has potential impact on the identification of differentially expressed genes. Segmentation is one of the important processes in image analysis. The current paper attempts to provide an overview of commonly used segmentation methods in microarray image analysis like fixed circle segmentation, adaptive circle segmentation, the adaptive shape segmentation, histogram-based method and machine learning algorithms. We estimated intensity ratios of selected spots from an image file downloaded from the Gene Expression Omnibus (GEO) database based on the above segmentation methods. It was observed that all these methods give almost similar estimates of intensity ratio value. We are also proposing a new algorithm to identify the spot radius for the adaptive circle segmentation, instead of manual fixing of the radius.
Keywords: Microarray, Image analysis, Segmentation, Intensity ratio.