Scanner-Agnostic MRI Harmonization via SSIM-Guided Disentanglement

TL;DR

This paper introduces a scanner-agnostic MRI harmonization method that disentangles anatomical features from scanner-specific variations using SSIM-guided loss, improving image consistency and downstream analysis accuracy.

Contribution

The proposed framework uniquely combines SSIM-guided disentanglement with multi-style harmonization for robust cross-site MRI normalization.

Findings

Achieved structural SSIM of 0.97 in harmonized images

Reduced Wasserstein distances between voxel intensity distributions

Improved brain age prediction error from 5.36 to 3.30 years

Abstract

The variability introduced by differences in MRI scanner models, acquisition protocols, and imaging sites hinders consistent analysis and generalizability across multicenter studies. We present a novel image-based harmonization framework for 3D T1-weighted brain MRI, which disentangles anatomical content from scanner- and site-specific variations. The model incorporates a differentiable loss based on the Structural Similarity Index (SSIM) to preserve biologically meaningful features while reducing inter-site variability. This loss enables separate evaluation of image luminance, contrast, and structural components. Training and validation were performed on multiple publicly available datasets spanning diverse scanners and sites, with testing on both healthy and clinical populations. Harmonization using multiple style targets, including style-agnostic references, produced consistent and high-quality outputs. Visual comparisons, voxel intensity distributions, and SSIM-based metrics demonstrated that harmonized images achieved strong alignment across acquisition settings while maintaining anatomical fidelity. Following harmonization, structural SSIM reached 0.97, luminance SSIM ranged from 0.98 to 0.99, and Wasserstein distances between mean voxel intensity distributions decreased substantially. Downstream tasks showed substantial improvements: mean absolute error for brain age prediction decreased from 5.36 to 3.30 years, and Alzheimer's disease classification AUC increased from 0.78 to 0.85. Overall, our framework enhances cross-site image consistency, preserves anatomical fidelity, and improves downstream model performance, providing a robust and generalizable solution for large-scale multicenter neuroimaging studies.