Spatially-informed Image Harmonization Results in Improved Scanner Effect Removal and Prediction

2026-04-01

AI Summary

This study addresses the challenge of technical variability introduced by different scanners in multi-site neuroimaging by proposing Tensor-ComBat, a novel method that explicitly models the spatial structure of voxels during batch-effect correction. Built upon Bayesian tensor response regression, Tensor-ComBat leverages low-rank PARAFAC decomposition to estimate spatially varying scanner effects and employs Markov chain Monte Carlo (MCMC) sampling to quantify uncertainty and enable rigorous statistical inference. Evaluated on over 2,100 T1-weighted MRI scans from the ADNI-1 cohort, the method demonstrates superior performance compared to existing approaches, effectively removing scanner-related artifacts while simultaneously enhancing image consistency, improving biological predictive accuracy, and increasing result reproducibility.

Abstract

We propose a novel data harmonization approach known as Tensor-ComBat (TC) for structural neuroimaging data. Tensor-Combat is a novel spatially aware harmonization method that aims to estimate and remove unwanted technical variation between voxel-level images from different study sites or scanners. Tensor-ComBat uses a Bayesian tensor response regression (BTRR) model to estimate spatially distributed scanner effects via a low-rank PARAFAC decomposition on the model coefficients, and subsequently removes these scanner effects via a post-hoc ComBat harmonization step. Unlike the classical ComBat method that treats the ROIs or voxels in the image as interchangeable, the Tensor-ComBat approach incorporates the information about the spatial configurations of imaging voxels when estimating the model parameters, resulting an improved harmonization pipeline. The proposed Tensor-ComBat method is fit using a Markov chain Monte Carlo (MCMC) that is not only able to produce point estimates, but also able to quantify uncertainty associated with these estimates. Due to the fully Bayesian implementation of Tensor-ComBat, it is able to perform inference related to significant spatially distributed scanner effects and/or biological effects that is not straightforward under existing methods. The methods is applied to over 2100 T1w-MRI scans in ADNI-1 that illustrates greater scanner effect removal, improved biological prediction, and superior reproducibility compared to state-of-the-art approaches such as ComBat.

Problem

• scanner effect
• image harmonization
• structural neuroimaging
• spatial variation
• technical variability

Innovation

• Tensor-ComBat
• spatially-aware harmonization
• Bayesian tensor regression
• scanner effect removal
• MCMC uncertainty quantification