Recently, classification and gene selection of DNA microarray data are important in biomedical research. DNA microarray data provide useful information that can be used to discover the complex mechanism of cancer development. DNA computing techniques are alternative approaches to analyze the DNA microarray data.
In this paper, we propose the VLSI implementation of a prostate cancer classification processor. The proposed architecture uses parallel and pipelined processing to improve speed and uses cyclic random masks to reduce memory size. We evaluated the prostate cancer classification processor by testing its performance on prostate cancer microarray data. From the experimental results, the proposed architecture reduced the memory size and classification time with little loss of classification accuracy.