SOFTWARE DEVELOPED FROM A FUZZY MATHEMATICAL MODEL TO PREDICT THE PATHOLOGICAL STAGE OF PROSTATE CANCER

In this work, we did the construction of a fuzzy mathematical model wich it was developed to predict the pathological stage of prostate cancer.⁵ The intention is to help the specialists on the decision process about stage of the disease, to avoid surgery and intensive treatments unnecessary. The model consists on a system founded in fuzzy rules, that it combine the pre-surgical data (clinic state, PSA level and Gleason score) availing of a set of linguistic rules made with base on informations of the existents nomograms. Herewith we hoped to get the chance of the individual, with certain clinical features, be in each stage of the tumor extension: localized, advanced locally and metastatic. Simulations were made with patient's data of the Clinics Hospital/UNICAMP and the results were compared with Kattan's probabilities⁸ that are used on the medicals decisions. A software was developed from this model and is a graphic interface that makes interaction with the subroutines that make the calculations. Its source code was written in Java and software has been tested on Windows and Linux / GNU.