Paper Details

The aim of this analysis was to develop a method for classification of cancers to specific diagnostic typesbased on their gene expression signature by applying Support Vector Machine (SVM).We trained the SVM by utilizingthe small, round blue-cell tumors (SRBCTs) as the model. These cancers belong to four distinct diagnostic categories andusually present diagnostic dilemmas in medical study. As their name implies, these cancers are difficult to distinguish bylight microscopy, and currently no single test can accurately distinguish these type of cancers. The SVM properlyclassified the whole samples and identified the genes most relevant to the classification. To test the ability of the trainedSVM models to identify SRBCTs, we examined additional blinded samples that were not previously used for the trainingpurpose, and correctly classified them in all cases. This study demonstrates the potential applications of these methodsfor tumor diagnosis and the identification of candidate targets for therapy. This paper presents architecture of SupportVector Machine classifiers arranged in a binary tree structure for solving multi-class classification problems withincreased efficiency.

Sangeeta Sharma, Kodanda Dhar Sa, “Classification and Diagnostic Prediction of Cancer Using Support Vector Machine”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 3, Issue 6, pp. 263-268, June 2016.