IDENTIFICATION OF EFFECTIVE GENES OF MULTIPLE CANCERS USING NEURAL NETWORK

Abstract

Cancer is a major health concern that affects a significant number of people worldwide and can often result in fatalities. Therefore, there is a growing need to develop effective approaches for early diagnosis and classification of different types of cancer. Early detection of cancer is crucial for prompt and accurate treatment. Thus, researchers have been working to identify non-invasive and precise methods for the early diagnosis, monitoring, and control of cancer. Leukemia and prostate cancer are two of the most common types of cancer globally. Microarray data analysis has become a valuable tool for diagnosing and classifying different types of cancerous tissues. To improve the accuracy of diagnosis, hybrid algorithms and neural networks are being employed. This paper provides a review of different biomarkers for leukemia and prostate cancer and proposes a novel method for distinguishing between the two cancers. The proposed method includes appropriate gene selection, a new hybrid model, and differential analysis of microarray data to create a diagnostic tool. The results indicate that the proposed algorithm is highly accurate and efficient in selecting a small set of valuable genes to improve classification accuracy. In conclusion, the accurate diagnosis and classification of cancer are essential for timely and effective treatment. The proposed method can contribute to the development of a reliable diagnostic tool for leukemia and prostate cancer, and the application of microarray data and hybrid algorithms can be useful for diagnosing other types of cancer as well.