Diagnosis Of Ovarian Cancer Using Artificial Neural Network

Ovarian cancer arises from the ovary indicates growth of cancer. More than 91%) ovarian cancers are known as "epithelial" which start from the surface epithelium, the thin tissue forming the outer layer of a body’s surface and lining the alimentary canal and other hollow structures of the ovary. The intention of this study is to observe the performance of the ANN algorithm over Genetic Algorithm on the diagnosis of ovarian cancer using proven ovarian dataset fig 2. Ovarian cancer [2] accounts for the most caused cancer diagnoses among women. We propose a comparison between Genetic Algorithm and ANN[1] for preoperative guess of enmity in ovarian tumors. Most of the present methods do not meet the requirements which deal with the drawbacks like accuracy and noise. Gene ranking methods like T-Score, ANOVA, went for wrong prediction the rank when large database is applied. The typical ANN is proposed to form part of a trustworthy tool to distinguish between kind and unkind ovarian tambours. This may help doctors to fix on the applicable treatment for the patients. Keywords--Gene Ranking, Trustworthy tool, ovarian tambours I .INTRODUCTION The diagnosis of complex genetic diseases like cancer has conventionally been done based on the non-molecular characteristics like kind of tumor tissue, pathological characteristics and clinical phase. Cancer precedents to almost 27% of all mortalities, making it the leading cause of death in America and also around the world. Timely and exact detection of cancer is life-threatening to the comfort of patients. Examinations of gene expression data precedents to cancer recognition and categorization, which will make ease appropriate treatment selection and drug development. A major application of microarrays has been to the study of cancer. Recognition of the signals that are symptoms for the disease phenotype and its progression requires the use of hardy techniques.Cancer can be identified through the analysis of genetic data. The human genome contains almost 10 million single nucleotide polymorphisms which will be in charge for the difference that lies between human beings. The microarray technology is used to achieve gene expression levels and SNPs of an individual. In this paper two methods are used namely Artificial Neural Network and Genetic algorithm and use techniques such as dimensionality reduction to improve the accuracy rate of the classifier .A genetic algorithm is a search experiment that reduces the process of natural selection. This experiment is regularly used to provide useful International Journal of Computer Trends and Technology (IJCTT) – volume 4 Issue10 – Oct 2013 ISSN: 2231-2803 http://www.ijcttjournal.org Page3602 solutions to enhance and search issues. Genetic algorithms [4] are appropriate to the huge class of evolutionary algorithms, which obtain solutions to enhancement problems using techniques that attract by natural progress, like heritage, alteration, collection, and boundary. Artificial neural networks are extensively used with appliances in science and technology An Artificial neural networks is a mathematical illustration of the human neural design, representing its “learning” and “generalization” capability. For this cause, Artificial neural networks have its place in the area of artificial intelligence. Artificial neural networks are extensively applied in research for the reason that they can model highly non-direct systems in which the relationship among the variables is undetermined or very complex.