Hardware Implementation of Hybrid Classifier to Detect Cancer Cells

Abstract

Convolutional Neural Network and Support Vector Machine are the common machine learning algorithms for classification. Both the classification techniques are fused to develop a hybrid classifier. The hardware implementation of hybrid classifier can improve the performance of the system and reduce the power consumption for real time applications. The main aim of this study is early detection of cancer cell using hybrid classifier implemented on a low cost handheld device. Leukemia and melanoma are the blood and skin cancer respectively. Both are dangerous form of cancers, many deaths have occurred due to leukemia and melanoma. In this paper, a hardware design is proposed to implement hybrid classifier on FPGA. The proposed system is implemented on Zync SoC FPGA platform and it gives the high performance, low power consumption and low hardware utilization.