Prostate Cancer Detection in MRI using Deep Learning

M. Hosseinzadeh

• Promotor: H. Huisman and B. van Ginneken
• Graduation year: 2024
• Radboud University, Nijmegen

Abstract

This thesis focuses on prostate cancer, a significant medical issue for men. The prostate is a component of the male reproductive system, comprising the penis, prostate, seminal vesicles, and testicles. Located below the bladder and in front of the rectum, the prostate is approximately the size of a walnut and encloses the urethra, the tube responsible for urine excretion. The prostate comprises distinct zones: the peripheral zone, which is commonly associated with prostate cancer, and the transition zone. The prostate produces fluid that contributes to semen composition. Prostate cancer ranks as the second most common cancer in men and the fourth most common cancer overall. Globally, over 1.4 million new cases of prostate cancer are diagnosed per year. Statistically, one in eight men will be diagnosed with prostate cancer in their lifetime, but only one in 41 will die as a result of the disease. This thesis aims to identify clinically significant prostate cancer (csPCa), which is curable if detected early.