Magnetic Resonance Imaging-based Prostate Cancer Detection and Classification using Machine Learning and Deep Learning Techniques: A Survey

Prostate cancer (PCa) ranks second among the most common cancers in men globally, with a staggering 1.4 million new cases estimated in 2020. Due to the increasing number of cases of this type of cancer, the detection and classification of PCa is one of the challenges for early treatment and increasing human survival chances. In recent years, several studies focused on identifying PCa. In the detection and classification of prostate cancer (PCa), magnetic resonance imaging (MRI) assumes a crucial role. Machine learning (ML) and Deep Learning (DL) can accurately and rapidly analyze MRI images and could provide better standardization and consistency in detecting PCa. This review summarizes studies that used ML and DL techniques on prostate MRI and focused on PCa detection and classification. The studies reviewed varied in performing prostate organ segmentation and approximation, prostate lesion characterization, and prostate lesion detection and segmentation. Derived from the good outcomes observed in the analysis of these studies, despite the difficulties and the presence of unsolved problems, it was concluded that the techniques based on ML and DL are adequate and promising for the detection and classification of PCa.