Medical Image Harmonization Using Deep Learning Based Canonical Mapping: Toward Robust and Generalizable Learning in Imaging
Abstract
Conventional and deep learning-based methods have shown great potential in the medical imaging domain, as means for deriving diagnostic, prognostic, and predictive biomarkers, and by contributing to precision medicine. However, these methods have yet to see widespread clinical adoption, in part due to limited generalization performance across various imaging devices, acquisition protocols, and patient populations. In this work, we propose a new paradigm in which data from a diverse range of acquisition conditions are "harmonized" to a common reference domain, where accurate model learning and prediction can take place. By learning an unsupervised image to image canonical mapping from diverse datasets to a reference domain using generative deep learning models, we aim to reduce confounding data variation while preserving semantic information, thereby rendering the learning task easier in the reference domain. We test this approach on two example problems, namely MRI-based brain age prediction and classification of schizophrenia, leveraging pooled cohorts of neuroimaging MRI data spanning 9 sites and 9701 subjects. Our results indicate a substantial improvement in these tasks in out-of-sample data, even when training is restricted to a single site.
Cite
@article{arxiv.2010.05355,
title = {Medical Image Harmonization Using Deep Learning Based Canonical Mapping: Toward Robust and Generalizable Learning in Imaging},
author = {Vishnu M. Bashyam and Jimit Doshi and Guray Erus and Dhivya Srinivasan and Ahmed Abdulkadir and Mohamad Habes and Yong Fan and Colin L. Masters and Paul Maruff and Chuanjun Zhuo and Henry Völzke and Sterling C. Johnson and Jurgen Fripp and Nikolaos Koutsouleris and Theodore D. Satterthwaite and Daniel H. Wolf and Raquel E. Gur and Ruben C. Gur and John C. Morris and Marilyn S. Albert and Hans J. Grabe and Susan M. Resnick and R. Nick Bryan and David A. Wolk and Haochang Shou and Ilya M. Nasrallah and Christos Davatzikos},
journal= {arXiv preprint arXiv:2010.05355},
year = {2020}
}