Intensity warping for multisite MRI harmonization

Abstract

In multisite neuroimaging studies there is often unwanted technical variation across scanners and sites. These "scanner effects" can hinder detection of biological features of interest, produce 1 inconsistent results, and lead to spurious associations. We assess scanner effects in two brain magnetic resonance imaging (MRI) studies where subjects were measured on multiple scanners within a short time frame, so that one could assume any differences between images were due to technical rather than biological effects. We propose mica (multisite image harmonization by CDF alignment), a tool to harmonize images taken on different scanners by identifying and removing within-subject scanner effects. Our goals in the present study were to (1) establish a method that removes scanner effects by leveraging multiple scans collected on the same subject, and, building on this, (2) develop a technique to quantify scanner effects in large multisite trials so these can be reduced as a preprocessing step. We found that unharmonized images were highly variable across site and scanner type, and our method effectively removed this variability by warping intensity distributions. We further studied the ability to predict intensity harmonization results for a scan taken on an existing subject at a new site using cross-validation.

Radiologist-Level Performance by Using Deep Learning for Segmentation of Breast Cancers on MRI Scans

“…Lateral resolution was left unchanged. Images from different scanners were harmonized by using intensity normalization, similar to previous work (37). For this purpose, a breast mask was established for each T1 image (see below).…”

Variability and Standardization of Quantitative Imaging

“…Harmonization can be performed using traveling human data acquired at each site by determining a scanner-specific correction factor. 40 – 42 If only postprocessed data (eg, fractional anisotropy map and cortical thickness) are available, regression analysis or more sophisticated statistical approaches can be performed for harmonization. 43 , 44 Harmonization of raw data is particularly important for diffusion-weighted imaging (DWI) data to be analyzed by multicompartment models or tractography, and model-free methods for harmonization of raw DWI signals have also been suggested.…”

Neurodevelopmental Patterns of Early Postnatal White Matter Maturation Represent Distinct Underlying Microstructure and Histology

“…While advances in fast and high-resolution acquisition of these basic MR sequences make it possible to resolve subtle signal changes of the newborn brain in the clinical setting (Kozak et al, 2020), one important technical hurdle is that MR acquisition across multiple sites/scanners inherently introduces undesired variations in image intensity and quality. Intensity harmonization techniques (Wrobel et al, 2020) or relaxometry methods (Leppert et al, 2009) can be utilized to diminish unwanted scanner effects and allow quantitative analysis and individuallevel inference for clinical applications. Second, we performed our study in MR images from healthy newborns without incidental neuroimaging findings of clinical significance.…”