Recognition of cDNA microarray image Using Feedforward artificial neural network

TL;DR

This paper introduces a neural network-based method for recognizing cDNA microarray images, improving gene identification accuracy through precise spot segmentation and ANN classification.

Contribution

It presents a novel approach combining spot segmentation and feedforward neural networks for cDNA microarray image recognition, enhancing matching performance.

Findings

Effective recognition of cDNA sequences demonstrated

Improved accuracy over previous methods

Neural network-based matching outperforms traditional techniques

Abstract

The complementary DNA (cDNA) sequence is considered to be the magic biometric technique for personal identification. In this paper, we present a new method for cDNA recognition based on the artificial neural network (ANN). Microarray imaging is used for the concurrent identification of thousands of genes. We have segmented the location of the spots in a cDNA microarray. Thus, a precise localization and segmenting of a spot are essential to obtain a more accurate intensity measurement, leading to a more precise expression measurement of a gene. The segmented cDNA microarray image is resized and it is used as an input for the proposed artificial neural network. For matching and recognition, we have trained the artificial neural network. Recognition results are given for the galleries of cDNA sequences . The numerical results show that, the proposed matching technique is an effective in the cDNA sequences process. We also compare our results with previous results and find out that, the proposed technique is an effective matching performance.