A Deep Learning Model Based on Convolutional Neural Networks for Classification of Magnetic Resonance Prostate Images

Abstract

When looking at prostate cancer, it is seen that it is one of the very common types of cancer in men. In literature review, it is understood that there are a lot of studies for the treatment and diagnosis of this type of cancer with various image processing methods on prostate images. On prostate biopsy, secondary haemorrhage areas of T2-weighted magnetic resonance (MR) in prostate images can cause false diagnoses. T1-weighted MR prostate images help diagnose these cases. In such cases, in order to prevent misdiagnosis; A new classification procedure for MR prostate images with convolutional neural networks (CNN) was performed. As a result of this process, a new deep learning model based on CNN which can classify T1-weighted and T2-weighted MR prostate images has been developed.