This work describes a simple and rapid method to detect a group of patients with prostate cancer from the acquisition of urine samples by implementing an electronic tongue composed of screen-printed carbon electrodes (C110). To validate the detection of prostate cancer samples, other samples were acquired from a group of patients called controls (i.e., healthy patients, patients with Benign Prostatic Hyperplasia and Prostatitis), which were subsequently classified using Pattern Recognition and Machine Learning methods. The results achieved were outstanding since it was possible to obtain up to 92.9% in the sample's classification accuracy.