@article{f5d1fddddb1f41b085d0a244f1781bf6,
title = "Deep Generative Medical Image Harmonization for Improving Cross-Site Generalization in Deep Learning Predictors",
abstract = "Background: In the medical imaging domain, deep learning-based methods have yet to see widespread clinical adoption, in part due to limited generalization performance across different imaging devices and acquisition protocols. The deviation between estimated brain age and biological age is an established biomarker of brain health and such models may benefit from increased cross-site generalizability. Purpose: To develop and evaluate a deep learning-based image harmonization method to improve cross-site generalizability of deep learning age prediction. Study Type: Retrospective. Population: Eight thousand eight hundred and seventy-six subjects from six sites. Harmonization models were trained using all subjects. Age prediction models were trained using 2739 subjects from a single site and tested using the remaining 6137 subjects from various other sites. Field Strength/Sequence: Brain imaging with magnetization prepared rapid acquisition with gradient echo or spoiled gradient echo sequences at 1.5 T and 3 T. Assessment: StarGAN v2, was used to perform a canonical mapping from diverse datasets to a reference domain to reduce site-based variation while preserving semantic information. Generalization performance of deep learning age prediction was evaluated using harmonized, histogram matched, and unharmonized data. Statistical Tests: Mean absolute error (MAE) and Pearson correlation between estimated age and biological age quantified the performance of the age prediction model. Results: Our results indicated a substantial improvement in age prediction in out-of-sample data, with the overall MAE improving from 15.81 (±0.21) years to 11.86 (±0.11) with histogram matching to 7.21 (±0.22) years with generative adversarial network (GAN)-based harmonization. In the multisite case, across the 5 out-of-sample sites, MAE improved from 9.78 (±6.69) years to 7.74 (±3.03) years with histogram normalization to 5.32 (±4.07) years with GAN-based harmonization. Data Conclusion: While further research is needed, GAN-based medical image harmonization appears to be a promising tool for improving cross-site deep learning generalization. Level of Evidence: 4. Technical Efficacy: Stage 1.",
keywords = "StarGAN, deep learning, harmonization",
author = "{The iSTAGING and PHENOM consortia} and Bashyam, {Vishnu M.} and Jimit Doshi and Guray Erus and Dhivya Srinivasan and Ahmed Abdulkadir and Ashish Singh and Mohamad Habes and Yong Fan and Masters, {Colin L.} and Paul Maruff and Chuanjun Zhuo and Henry V{\"o}lzke and Johnson, {Sterling C.} and Jurgen Fripp and Nikolaos Koutsouleris and Satterthwaite, {Theodore D.} and Wolf, {Daniel H.} and Gur, {Raquel E.} and Gur, {Ruben C.} and Morris, {John C.} and Albert, {Marilyn S.} and Grabe, {Hans J.} and Resnick, {Susan M.} and Bryan, {Nick R.} and Katharina Wittfeld and Robin B{\"u}low and Wolk, {David A.} and Haochang Shou and Nasrallah, {Ilya M.} and Christos Davatzikos",
note = "Funding Information: This work was supported in part by NIH grants RF1AG054409, R01EB022573, R01MH112070, R01MH120482, R01MH113565, and NIH contract HHSN271201600059C. This work was also supported in part by the Intramural Research Program, National Institute on Aging, NIH, and the Swiss National Science Foundation grant 191026. SHIP is part of the Community Medicine Research net of the University of Greifswald, Germany, which is funded by the Federal Ministry of Education and Research (grants 01ZZ9603, 01ZZ0103, and 01ZZ0403), the Ministry of Cultural Affairs, and the Social Ministry of the Federal State of Mecklenburg-West Pomerania. MRI scans in SHIP and SHIP-TREND have been supported by a joint grant from Siemens Healthineers, Erlangen, Germany and the Federal State of Mecklenburg-West Pomerania. Funding Information: This work was supported in part by NIH grants RF1AG054409, R01EB022573, R01MH112070, R01MH120482, R01MH113565, and NIH contract HHSN271201600059C. This work was also supported in part by the Intramural Research Program, National Institute on Aging, NIH, and the Swiss National Science Foundation grant 191026. SHIP is part of the Community Medicine Research net of the University of Greifswald, Germany, which is funded by the Federal Ministry of Education and Research (grants 01ZZ9603, 01ZZ0103, and 01ZZ0403), the Ministry of Cultural Affairs, and the Social Ministry of the Federal State of Mecklenburg‐West Pomerania. MRI scans in SHIP and SHIP‐TREND have been supported by a joint grant from Siemens Healthineers, Erlangen, Germany and the Federal State of Mecklenburg‐West Pomerania. Publisher Copyright: {\textcopyright} 2021 International Society for Magnetic Resonance in Medicine.",
year = "2022",
month = mar,
doi = "10.1002/jmri.27908",
language = "English (US)",
volume = "55",
pages = "908--916",
journal = "Journal of Magnetic Resonance Imaging",
issn = "1053-1807",
publisher = "John Wiley and Sons Inc.",
number = "3",
}