A Hybrid Kernel Extreme Learning Machine based Approach for Microarray Gene Expression and Multiclass Cancer Classification

Karthikeyan, T.; Balakrishnan, R.
May 2014
Australian Journal of Basic & Applied Sciences;May2014, Vol. 8 Issue 7, p408
Academic Journal
Nowadays cancer is one of the most dreadful diseases, which causes a substantial death rate in humans. Cancer is characteristic by an irregular, uncontrollable growth that may demolish and attack nearest healthy body tissues or somewhere else in the body. Microarray based gene expression profiling has been come out as an efficient technique for cancer classification, analysis, prognosis and for treatment purposes. Recently, DNA microarray technique has got more attention in both scientific and in industrial fields. It showed great importance in deceiving the informative genes that can cause the cancer. This led to improvements in early on cancer diagnosis and in giving effective treatment. The multicategory cancer classification plays an important role in the field of medical sciences. As the numbers of cancer victims are increasing gradually, the requirement of the cancer classification system will become a necessary one. For the above impenetrability and to obtain better consequences of the system with accuracy a combination of Integer-Coded Genetic Algorithm (ICGA) and a Hybrid Particle Swarm Optimization with Artificial Bee Colony (HPSABC) model is proposed. HPSABC algorithm is a combination of Particle Swarm Algorithm (PSO) and Artificial Bee Colony (ABC) Algorithm. The approach is mainly used to find the optimal gene selection that should be assigned in order to attain maximum cost effectiveness. This HPSABC algorithm is coupled with an extreme learning machine based on hybrid kernel function (HKELM), is used for gene selection and cancer classification. ICGA is used with HPSABC- HKELM to select an optimal set of genes, which is then used to build a classifier to develop an algorithm (ICGA_ HPSABC_ HKELM) that can handle sparse data and sample imbalance data. The performance of the proposed algorithm is evaluated and compares the results with existing methods in the literature.


Related Articles

  • Microarray Gene Expression and Multiclass Cancer Classification Using Improved PSO Based Evolutionary Fuzzy ELM Classifier with ICGA Gene Selection. Karthikeyan, T.; Balakrishnan, R. // International Review on Computers & Software;Oct2013, Vol. 8 Issue 10, p2532 

    Cancer has become one of the dreadful and most widely spreading diseases in recent years. Cancer diagnosis has become an active research area in the field of medical image processing. DNA microarrays are considered as an effectual tool used in molecular biology and cancer diagnosis. As the...

  • Greedy Search-Binary PSO Hybrid for Biclustering Gene Expression Data.  // International Journal of Computer Applications;Mar2010, Vol. 2, p1 

    The article offers information on the capability of a data mining technique to identify local patterns from gene expression data. It presents a new method based on greedy search algorithm combined with particle swarm optimization for the identification of biclusters in gene expression data. It...

  • Improving prediction accuracy of tumor classification by reusing genes discarded during gene selection. Yang, Jack Y.; Guo-Zheng Li; Hao-Hua Meng; Qu Yang, Mary; Youping Deng // BMC Genomics;2008 Supplement 1, Vol. 9, Special section p1 

    Background: Since the high dimensionality of gene expression microarray data sets degrades the generalization performance of classifiers, feature selection, which selects relevant features and discards irrelevant and redundant features, has been widely used in the bioinformatics field. Multitask...


    With the advent of microarray technology, it is possible to measure gene expression levels of thousands of genes simultaneously. This helps us diagnose and classify some particular cancers directly using DNA microarray. Highdimensionality and small sample size of microarray datasets has made the...

  • Human Liver Cancer Classification using Microarray Gene Expression Data. Rajeswari, P.; Reena, G. Sophia // International Journal of Computer Applications;Nov2011, Vol. 34, p25 

    Cancer is one of the dreadful diseases, which causes considerable death rate in humans. Cancer is featured by an irregular, unmanageable growth that may demolish and attack neighboring healthy body tissues or somewhere else in the body. There are dissimilar techniques lives for the naming of...

  • NIM: A Node Influence Based Method for Cancer Classification. Yiwen Wang; Min Yao; Jianhua Yang // Computational & Mathematical Methods in Medicine;Jan2014, p1 

    The classification of different cancer types owns great significance in the medical field. However, the great majority of existing cancer classification methods are clinical-based and have relatively weak diagnostic ability. With the rapid development of gene expression technology, it is able to...

  • LEAF: Leave-one-out Forward Selection Method for Informative Gene Discovery in DNA Microarray Data. Fukuta, Kentaro; Okada, Yoshifumi // IAENG International Journal of Computer Science;Jun2011, Vol. 38 Issue 2, p160 

    Preventing, diagnosing, and treating disease is greatly facilitated by the availability of biomarkers. Recent improvements in bioinformatics technology have facilitated large-scale screening of DNA microarrays for candidate biomarkers. Here we discuss a gene selection method, which is called...

  • Gene selection using independent variable group analysis for tumor classification. Zheng, Chun-Hou; Chong, Yan-Wen; Wang, Hong-Qiang // Neural Computing & Applications;Mar2011, Vol. 20 Issue 2, p161 

    Microarrays are capable of detecting the expression levels of thousands of genes simultaneously. So, gene expression data from DNA microarray are characterized by many measured variables (genes) on only a few samples. One important application of gene expression data is to classify the samples....

  • Decision Tree Classifiers in Bioinformatics. Polaka, Inese // Computer Science (1407-7493);2010, Vol. 44, p119 

    No abstract available.


Read the Article


Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics