Capsule GAN for Prostate MRI Super-Resolution

TL;DR

This paper introduces a novel Capsule GAN model for prostate MRI super-resolution, significantly improving image quality and diagnostic accuracy, aiding early prostate cancer detection.

Contribution

A new Capsule GAN-based super-resolution model for prostate MRI that outperforms existing methods and includes a task-specific similarity assessment for medical detail reconstruction.

Findings

Outperforms state-of-the-art prostate SR models in all similarity metrics.

Introduces a task-specific similarity assessment for medical images.

Demonstrates potential for improved early prostate cancer diagnosis.

Abstract

Prostate cancer is a very common disease among adult men. One in seven Canadian men is diagnosed with this cancer in their lifetime. Super-Resolution (SR) can facilitate early diagnosis and potentially save many lives. In this paper, a robust and accurate model is proposed for prostate MRI SR. The model is trained on the Prostate-Diagnosis and PROSTATEx datasets. The proposed model outperformed the state-of-the-art prostate SR model in all similarity metrics with notable margins. A new task-specific similarity assessment is introduced as well. A classifier is trained for severe cancer detection and the drop in the accuracy of this model when dealing with super-resolved images is used for evaluating the ability of medical detail reconstruction of the SR models. The proposed SR model is a step towards an efficient and accurate general medical SR platform.