DA-VSR: Domain Adaptable Volumetric Super-Resolution For Medical Images

TL;DR

DA-VSR introduces a domain-adaptive volumetric super-resolution method for medical images, combining supervised feature learning with unsupervised domain adaptation to enhance image quality across diverse datasets.

Contribution

It proposes a novel framework that effectively bridges domain gaps in medical image super-resolution using a unified backbone and self-learned domain adaptation techniques.

Findings

Significantly improves super-resolution quality across multiple datasets.

Effectively adapts to domain differences using self-learned unsupervised learning.

Enhances clinical applicability of medical image super-resolution.

Abstract

Medical image super-resolution (SR) is an active research area that has many potential applications, including reducing scan time, bettering visual understanding, increasing robustness in downstream tasks, etc. However, applying deep-learning-based SR approaches for clinical applications often encounters issues of domain inconsistency, as the test data may be acquired by different machines or on different organs. In this work, we present a novel algorithm called domain adaptable volumetric super-resolution (DA-VSR) to better bridge the domain inconsistency gap. DA-VSR uses a unified feature extraction backbone and a series of network heads to improve image quality over different planes. Furthermore, DA-VSR leverages the in-plane and through-plane resolution differences on the test data to achieve a self-learned domain adaptation. As such, DA-VSR combines the advantages of a strong feature generator learned through supervised training and the ability to tune to the idiosyncrasies of the test volumes through unsupervised learning. Through experiments, we demonstrate that DA-VSR significantly improves super-resolution quality across numerous datasets of different domains, thereby taking a further step toward real clinical applications.