Abstract
Our knowledge of genomic medicine has increased at astonishing pace in recent years. As a result of advances in genomics, proteomics and molecular pathology, many candidate biomarkers with potential clinical value have been identified. Furthermore, recent refinements of experimental technique including array-based assays which generate huge datasets and find numerous potentially useful molecules are becoming widespread. The main aim of this research is to develop and validate an intelligent modelling system for the analysis of large array datasets in a nonlinear manner in order to diagnose and predict the bladder cancer disease. In this work, an investigation of epigenetic data for the diagnosis and progression of UCC was performed using intelligent firefly system, based on various clinicopathological criteria of CpG and involving tumour behaviour. An introduction of the most common artificial intelligent-based modelling techniques and gene expression data predictive modelling will be carried out. Experimental work based on MATLAB neural network toolbox will be described and discussed. A new module based on ANN analyse a large array datasets of human genes which can predict the diagnosis and progression of cancer bladder and help to discover the genes causing cancer. Experimental data was collected using intelligent firefly system and tested with artificial neural network technique.