Artificial Intelligence Technique for Gene Expression Profiling of Urinary Bladder Cancer

Abstract

The purpose of this study is to develop a method of classifying cancers to specific diagnostic categories based on their gene expression signatures using artificial intelligence (AI) techniques which provide better predictions than standard traditional statistical methods. The predictive accuracies of neuro-fuzzy modelling (NFM), artificial neural networks (ANN) and traditional logistic regression (LR) methods are compared for the behaviour of bladder cancer. Gene expression profiles of non-invasive and invasive bladder cancer were used to identify potential therapeutic or screening targets in bladder cancer, and to define genetic changes relevant for tumour progression of recurrent papillary bladder cancer (pTa). For all three methods, models were produced to predict the presence and timing of a tumour progression, stage and grade. AI methodology predicted progression with an accuracy ranging up to 100%. This was superior to logistic regression