Deep learning in prostate cancer diagnosis and Gleason grading in histopathology images: An extensive study

• Main Authors: Ali Hasan Md. Linkon, Md. Mahir Labib, Tarik Hasan, Mozammal Hossain, Marium-E- Jannat
• Format: Article
• Language: English
• Published: Elsevier 2021-01-01
• Series: Informatics in Medicine Unlocked
• Subjects: Deep learning; Convolutional neural network; Computer-aided detection; Medical imaging; Prostate cancer detection; Gleason grading
• Online Access: http://www.sciencedirect.com/science/article/pii/S2352914821000721

Among American men, prostate cancer is the cause of the second-highest death by any cancer. It is also the most common cancer in men worldwide, and the annual numbers are quite alarming. The most prognostic marker for prostate cancer is the Gleason grading system on histopathology images. Pathologists determine the Gleason grade on stained tissue specimens of Hematoxylin and Eosin (H&E) based on tumor structural growth patterns from whole slide images. Recent advances in Computer-Aided Detection (CAD) using deep learning have brought the immense scope of automatic detection and recognition at very high accuracy in prostate cancer like other medical diagnoses and prognoses. Automated deep learning systems have delivered promising results from histopathological images to accurate grading of prostate cancer. Many studies have shown that deep learning strategies can achieve better outcomes than simpler systems that make use of pathology samples. This article aims to provide an insight into the gradual evolution of deep learning in detecting prostate cancer and Gleason grading. This article also evaluates a comprehensive, synthesized overview of the current state and existing methodological approaches as well as unique insights in prostate cancer detection using deep learning. We have also described research findings, current limitations, and future avenues for research. We have tried to make this paper applicable to deep learning communities and hope it will encourage new collaborations to create dedicated applications and improvements for prostate cancer detection and Gleason grading.