Abstract

The process of diagnosing of prostate cancer using traditional methods is cumbersome because of the similarity of symptoms that are present in other diseases. Soft Computing (SC) paradigms which mimic human imprecise data manipulation and learning capabilities have been reviewed and harnessed for diagnosis and classification of prostate cancer. SC technique based on Adaptive Neuro-Fuzzy Inference System (ANFIS) facilitated symptoms analysis, diagnosis and prostate cancer classification. Age of Patient (AP), Pains in Urination (PU), Frequent Urination (FU), Blood in Semen (BS) and Pains in Pelvic (PP) served as input attributes while Prostate Risk (PR) served as output. Matrix laboratory provided the programming tools for system implementation. The practical function of the system was assessed using prostate cancer data collected from the University of Uyo Teaching Hospital. A 95% harmony observed between the computed and the expected output in the ANFIS model, showed the superiority of the ANFIS model over the fuzzy model. The system is poised to assist medical professionals in the domain of diagnosis and classification of prostate cancer for the promotion of management and treatment decisions.