Prostate Cancer Diagnosis Automation Using Supervised Artificial Intelligence. A Systematic Literature Review

Abstract

Prostate cancer (PCa) is the most frequent genre-specific malignancy and the fourth overall behind lung, breast, and colon cancers. PCa is diagnosed non-invasively with serum prostate-specific antigen assay, digital rectal examination, and trans-rectal ultrasound and invasively with multiple rectal biopsies from which a Gleason score is assigned. The biopsy tissue is subject to sampling error and cellular interpretation that in turn can lead to disagreement as to whether treatment is needed, and if so, the method and the extent of therapy. Magnetic resonance (MR) imaging is proving to be progressively useful in evaluating PCa. New sequences are continually being introduced that are proving to be even more accurate in determining the extent and degree of tumor malignancy than other imaging modalities. The MR images, however, are evaluated by radiologists whose interpretation is subjective. This study reviews the currently available artificial intelligence and machine learning techniques that may eliminate the need for multiple rectal biopsies and provide a more uniform classification of these malignancies. Also, the evaluation of treatment outcome can be better assessed with more precise tumor size and classification. This paper investigates and analyzes projects related to prostate cancer’s automatic diagnosis using artificial intelligence.