Robust diffusion imaging framework for clinical studies

TL;DR

This paper presents a robust diffusion imaging framework that effectively restores and utilizes corrupted clinical MRI datasets, improving reliability and efficiency in medical imaging analysis despite artefacts and low signal quality.

Contribution

The authors developed an improved least trimmed squares diffusion tensor estimation algorithm that handles severely degraded datasets, enabling the use of corrupted images in clinical diffusion studies.

Findings

The framework outperforms existing algorithms in simulations and in vivo tests.

Corrupted datasets can be effectively restored and reused in clinical analyses.

The method is applicable to other MR studies requiring artefact suppression.

Abstract

Clinical diffusion imaging requires short acquisition times and good image quality to permit its use in various medical applications. In turn, these demands require the development of a robust and efficient post-processing framework in order to guarantee useful and reliable results. However, multiple artefacts abound in in vivo measurements; from either subject such as cardiac pulsation, bulk head motion, respiratory motion and involuntary tics and tremor, or imaging hardware related problems, such as table vibrations, etc. These artefacts can severely degrade the resulting images and render diffusion analysis difficult or impossible. In order to overcome these problems, we developed a robust and efficient framework enabling the use of initially corrupted images from a clinical study. At the heart of this framework is an improved least trimmed squares diffusion tensor estimation algorithm that works well with severely degraded datasets with low signal-to-noise ratio. This approach has been compared with other diffusion imaging post-processing algorithms using simulations and in vivo experiments. Exploiting track-based spatial statistics analysis, we demonstrate that corrupted datasets can be restored and reused in further clinical studies rather than being discarded due to poor quality. The developed robust framework is shown to exhibit a high efficiency and accuracy and can, in principle, be exploited in other MR studies where artefact/outlier suppression is needed.