SVM Classification of Cell Survival/Apoptotic Death for Color Texture Images of Survival Receptor Proteins

Abstract

In recent years, considerable interests have been developed for the development of small cell inhibitors of protein kinases. Different combinations are undertaken human trials for the medical care of cancer, inflammatory diseases, and other symptoms, while some of them are validated for clinical purpose only. Apoptosis and necrosis are two different types of cell death. In necrosis, serious health problems and inflammatory responses occurs due to uncontrolled cell death while in apoptosis there is a control on cell death. Cancer arises from the dysfunction in the apoptotic pathway. Apoptosis or cell death is a general component for the evolution of multicellular organisms. In this paper a CAD system is designed for the classification (survival or apoptotic death) of survival receptor proteins (EGF and Insulin) using Discrete Wavelet transform for colour texture classification problem. In this article all the work has been done on the color images which has not done till yet by other researchers. The maximum accuracy of 93.33% is obtained using Support vector machine (SVM) classifier while the minimum accuracy of 13.33% is obtained.