AUTHOR=Kim SeungWook , Kim Sung-Woo , Noh Young , Lee Phil Hyu , Na Duk L. , Seo Sang Won , Seong Joon-Kyung 

TITLE=Harmonization of Multicenter Cortical Thickness Data by Linear Mixed Effect Model

JOURNAL=Frontiers in Aging Neuroscience

VOLUME=Volume 14 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/aging-neuroscience/articles/10.3389/fnagi.2022.869387

DOI=10.3389/fnagi.2022.869387

ISSN=1663-4365

ABSTRACT=Objective: Analyzing neuroimages being useful method in the field of neuroscience and neurology, solving the incompatibilities across protocols and vendors have become a major problem. We refer to this incompatibility as “center effects”, and in this study we attempt to remove such center effects of cortical feature obtained from multi-center magnetic resonance images (MRI).
Methods: For MRI of a total of 4321 multi-center subjects, harmonized w-score was calculated by correcting biological covariates such as age, sex, years of education, and intercranial volume (ICV) as fixed effects and center information as random effect. Afterwards, we performed classification tasks using principal component analysis (PCA) and linear discriminant analysis (LDA) to check whether the center effect was successfully removed from the harmonized w-score.
Results: First, an experiment was conducted to predict the dataset origin of a random subject sampled from two different datasets, and it was confirmed that prediction accuracy of LME model-based w-score was significantly closer to the baseline than that of raw cortical thickness. As a second experiment, we classified the data of the normal and patient groups of each dataset, and LME model-based w-score, which is biological-feature-corrected values, showed higher classification accuracy than the raw cortical thickness data. Afterwards, in order to verify the compatibility of the dataset used for LME model training and the dataset that is not, intra-object comparison and w-score RMSE calculation process were performed.
Conclusion: Through comparison between LME model-based w-score and existing methods and several classification tasks, we showed that LME model-based w-score successfully removes the center effects while preserving the disease effects from the dataset. We also show that the preserved disease effects have a match with well-known disease atrophy patterns such as Alzheimer’s disease or Parkinson’s disease. Finally, through intra-subject comparison, we found that the difference between centers decreases in the LME model-based w-score compared to the raw cortical thickness, and thus showed that our model well-harmonizes the data which is not used for the model training.