Computer Aided Diagnosis System for Prediction of Gleason Score in Prostate Cancer

Prostate Cancer (PCa) is the most common type of cancer in males and the third most diagnosed cancer overall inthe world. Early Detection of PCa increases the chances of successfully treating it. Present detection methods like transrectal ultrasound (TRUS) biopsy are invasive while Prostate-Specific Antigen (PSA) tests have high false positive rates. Magnetic Resonance Imaging (MRI) has revolutionized the way prostate cancer is detected with the help of Computer-Aided Diagnosis (CAD). This research work developed a fully automated end-to-end detection and grading system with mpMRI from Prostate X-2 dataset which was used to detect the cancerous lesions and grade cancer aggressiveness according to Gleason Grade Groups (GGG) scores. This study developed an end-to-endsystem with a VGG19 feature extraction model to extract features, which were fed into a Random Forest classifier for the classification of Gleason Grade Groups. This system can assist healthcare professionals in the early detection of PCa lesions and classifying the aggressiveness using the GGG scores. Additional research in this field has the potential to advance both healthcare technology and improve patient care.