PT  - JOURNAL ARTICLE
AU  - Lijun An
AU  - Jianzhong Chen
AU  - Pansheng Chen
AU  - Tong He
AU  - Christopher Chen
AU  - Juan Helen Zhou
AU  - B.T. Thomas Yeo
AU  - the Alzheimer’s Disease Neuroimaging Initiative
AU  - the Australian Imaging Biomarkers and Lifestyle Study of Aging
TI  - Goal-specific brain MRI harmonization
AID  - 10.1101/2022.03.05.483077
DP  - 2022 Jan 01
TA  - bioRxiv
PG  - 2022.03.05.483077
4099  - http://biorxiv.org/content/early/2022/03/07/2022.03.05.483077.short
4100  - http://biorxiv.org/content/early/2022/03/07/2022.03.05.483077.full
AB  - There is significant interest in pooling magnetic resonance image (MRI) data from multiple datasets to enable mega-analysis. Harmonization is typically performed to reduce heterogeneity when pooling MRI data across datasets. Most MRI harmonization algorithms do not explicitly consider downstream application performance during harmonization. However, the choice of downstream application might influence what might be considered as study-specific confounds. Therefore, ignoring downstream applications during harmonization might potentially limit downstream performance. Here we propose a goal-specific harmonization framework that utilizes downstream application performance to regularize the harmonization procedure. Our framework can be integrated with a wide variety of harmonization models based on deep neural networks, such as the recently proposed conditional variational autoencoder (cVAE) harmonization model. Three datasets from three different continents with a total of 2787 participants and 10085 anatomical T1 scans were used for evaluation. We found that cVAE removed more dataset differences than the widely used ComBat model, but at the expense of removing desirable biological information as measured by downstream prediction of mini mental state examination (MMSE) scores and clinical diagnoses. On the other hand, our goal-specific cVAE (gcVAE) was able to remove as much dataset differences as cVAE, while improving downstream cross-sectional prediction of MMSE scores and clinical diagnoses.Competing Interest StatementThe authors have declared no competing interest.