Oversampling errors in multimodal medical imaging are due to the Gibbs effect

TL;DR

This paper investigates the causes of oversampling errors in multimodal 3D medical imaging, identifying the Gibbs effect as a key contributor, and compares resampling strategies across different neuroimaging tools.

Contribution

It provides an analysis of oversampling errors in medical image resampling, highlighting the Gibbs effect's role and recommending undersampling for improved accuracy.

Findings

Oversampling errors are larger in regions with steeper gradients.

Undersampling to the lowest image size reduces mean segment errors.

The Gibbs effect significantly contributes to oversampling errors.

Abstract

To analyse multimodal 3-dimensional medical images, interpolation is required for resampling which - unavoidably - introduces an interpolation error. In this work we consider three segmented 3-dimensional images resampled with three different neuroimaging software tools for comparing undersampling and oversampling strategies and to identify where the oversampling error lies. The results indicate that undersampling to the lowest image size is advantageous in terms of mean value per segment errors and that the oversampling error is larger where the gradient is steeper, showing a Gibbs effect.