Cancer Detection Based on Classification Microarray Data

Posted by JMII e-Journal on 18.58

Cancer Detection Based on Classification Microarray Data Using PCA and Modified Back Propagation


Adiyasa Nurfalah 1), Adiwijaya 2), Arie Ardiyanti Suryani 3)
Telkom University
St. Telekomunikasi No. 1, Bandung, Indonesia
E-mail : adiyasa.nurfalah@gmail.com 1), kang.ady@gmail.com 2), rie006@yahoo.com 3)



AbstractCancer is the leading cause of death in the world based on data from the World Health Organization (WHO) in 2012, which is about 8.2 million die because cancer and estimated will increase each year due to an unhealthy lifestyle [2]. Deaths due to cancer could be prevented if the cancer was detected early. In recent decades microarray has taken an important role in cancer research.
Microarray is a technology that is capable of storing thousands of gene expressions taken from several specific tissues of human at once. By analyzing microarray data can be known some affected by cancer or not. In this study built a fast and accurate framework for cancer detection based on microarray data classification using principal component analysis (PCA) and modified back propagation (MBP). MBP is a modification of standard back propagation (BP) by implementing conjugate gradient algorithm on search direction in BP training. Generally, conjugate gradient algorithm required shorter training time than steepest descent that used by standard back propagation.
The experiment results was showed MBP-based system (PCA+MBP or MBP) is able to outperform BP-based system (BP or PCA+BP) in accuracy and especially in training time.
Keywords—cancer detection, microarray, principal component analysis, conjugate gradient algorithm, back propagation.

Paper 1 JMII Vol 1  No 1 Tahun  2016