Free water DTI estimates from single b-value data might seem plausible but must be interpreted with care

TL;DR

This study evaluates the accuracy of free water elimination diffusion tensor imaging (FWE-DTI) methods, especially the regularized gradient descent (RGD), on single-shell data, highlighting its limited specificity and recommending multi-shell data and NLS methods for better results.

Contribution

The paper critically assesses the validity of RGD FWE-DTI on single-shell data and compares it with NLS, emphasizing the importance of acquisition type and initialization in parameter estimation.

Findings

RGD can produce plausible maps from single-shell data but lacks specificity.

NLS method can distinguish free water from tissue diffusion alterations.

Multi-shell data and NLS are recommended for accurate FWE-DTI analysis.

Abstract

Purpose: Free water elimination diffusion tensor imaging (FWE-DTI) has been widely used to distinguish increases of free water (FW) partial volume effects from tissue's diffusion in healthy ageing and degenerative diseases. Since the FWE-DTI fitting is only well posed for multi-shell acquisitions, a regularized gradient descent (RGD) method was proposed to enable application to single-shell data, more common in the clinic. However, the validity of the RGD method has been poorly assessed. This study aims to quantify the specificity of FWE-DTI procedures on single- and multi-shell data. Methods: Different FWE-DTI fitting procedures were tested on an open-source in vivo diffusion dataset and single- and multi-shell synthetic signals, including the RGD and standard non-linear least squares (NLS) methods. Single-voxel simulations were carried out to compare initialization approaches. A multi-voxel phantom simulation was performed to evaluate the impact of spatial regularization when comparing between methods. To test the algorithms' specificity, phantoms with two different types of lesions were simulated: with increased mean diffusivity (MD) or with increased FW. Result: Plausible parameter maps were obtained with RGD from single-shell in vivo data. The plausibility of these maps was shown to be determined by the initialization. Tests with simulated lesions inserted into the in vivo data revealed that the RGD approach cannot distinguish FW from tissue MD alterations, contrarily to the NLS algorithm. Conclusion: RGD FWE-DTI has limited specificity and, thus, its results from single-shell data should be carefully interpreted. When possible, multi-shell acquisitions and the NLS approach should be preferred instead.