Genetic Information Gain based on Optimized Multilayer Neural Network Techniques for Early Identification of Colon Cancer

Colon Cancer (CC) is the third leading cause of cancer-related deaths worldwide. It is a global public health problem that must be addressed in diagnosis and treatment to repository of thousands of gene expression patterns that vary in intensity from one cancer cell to another. Despite the previous methods of detecting this cancer, they do not detect the profound aspects of cancer, and thus, early detection of CC is challenging. There is a need to find faster and more precise tissue identification methods and improve the diagnostic process. To combat these challenges, this novel presents two-stage, Genetic Information Gain (GIG) and Optimized Multilayer Neural Network (OMNN) approaches to enhance the identification of CC. To begin with, data preprocessing using the Box-plot Normalization process (Bp-Np) method. Subsequently, the Synthetic Minority Oversampling Technique (SMOsT) approach is employed to analyze the CC impact margins from the processed dataset. Moreover, the Colon Cancer Periodic Influence Rate (CCPIR) method is used to find the criteria’s weight. Furthermore, the GIG approach picks profound aspects of CC. Lately, we have employed an MNN optimized with a linear vectorized model to identify colon cancer based on its in-depth features for early detection. Hence, it is concluded that the proposed technique's performance gains higher accuracy than existing methods.