Gel Image Matching Using Hopfield Neural Network


The KIPS Transactions:PartB , Vol. 13, No. 3, pp. 323-328, Jun. 2006
10.3745/KIPSTB.2006.13.3.323,   PDF Download:

Abstract

Proteins in a cell appear as spots in a two dimensional gel image which is used in protein analysis. The spots from the same protein are in near position when comparing two gel images. Finding out the different proteins between a normal tissue and a cancer one is important information in drug development. Automatic matching of gel images is difficult because they are made from biological experimental processes. This matching problem is known to be NP-hard. Neural networks are usually used to solve such NP-hard problems. Hopfield neural network is selected since it is appropriate to solve the gel matching. An energy function with location and distance parameters is defined. The two spots which make the energy function minimum are matching spots and they came from the same protein. The energy function is designed to reflect the topology of spots by examining not only the given spot but also neighborhood spots.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
Y. Ankhbayar, S. H. Hwang, Y. S. Hwang, "Gel Image Matching Using Hopfield Neural Network," The KIPS Transactions:PartB , vol. 13, no. 3, pp. 323-328, 2006. DOI: 10.3745/KIPSTB.2006.13.3.323.

[ACM Style]
Yukhuu Ankhbayar, Suk Hyung Hwang, and Young Sup Hwang. 2006. Gel Image Matching Using Hopfield Neural Network. The KIPS Transactions:PartB , 13, 3, (2006), 323-328. DOI: 10.3745/KIPSTB.2006.13.3.323.