Enhance the Image: Super Resolution using Artificial Intelligence in MRI

TL;DR

This paper reviews deep learning methods for MRI super-resolution, analyzing their architectures, evaluation metrics, and clinical impact, while discussing challenges and future directions for wider adoption.

Contribution

It provides a comprehensive overview of advanced deep learning models for MRI super-resolution and evaluates their clinical and neuroscientific implications.

Findings

Deep learning models improve MRI resolution effectively.

Evaluation metrics are crucial for assessing super-resolution quality.

Challenges include data variability and model reliability.

Abstract

This chapter provides an overview of deep learning techniques for improving the spatial resolution of MRI, ranging from convolutional neural networks, generative adversarial networks, to more advanced models including transformers, diffusion models, and implicit neural representations. Our exploration extends beyond the methodologies to scrutinize the impact of super-resolved images on clinical and neuroscientific assessments. We also cover various practical topics such as network architectures, image evaluation metrics, network loss functions, and training data specifics, including downsampling methods for simulating low-resolution images and dataset selection. Finally, we discuss existing challenges and potential future directions regarding the feasibility and reliability of deep learning-based MRI super-resolution, with the aim to facilitate its wider adoption to benefit various clinical and neuroscientific applications.