Literature DB >> 34564904

Deep Generative Medical Image Harmonization for Improving Cross-Site Generalization in Deep Learning Predictors.

Vishnu M Bashyam1, Jimit Doshi1, Guray Erus1, Dhivya Srinivasan1, Ahmed Abdulkadir1, Ashish Singh1, Mohamad Habes2, Yong Fan1, Colin L Masters3, Paul Maruff3, Chuanjun Zhuo4,5, Henry Völzke6,7, Sterling C Johnson8, Jurgen Fripp9, Nikolaos Koutsouleris10, Theodore D Satterthwaite1,11, Daniel H Wolf11, Raquel E Gur11,12, Ruben C Gur11,12, John C Morris13, Marilyn S Albert14, Hans J Grabe15,16, Susan M Resnick17, Nick R Bryan18, Katharina Wittfeld15,16, Robin Bülow19, David A Wolk20, Haochang Shou21, Ilya M Nasrallah12, Christos Davatzikos1.   

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.
© 2021 International Society for Magnetic Resonance in Medicine.

Entities:  

Keywords:  StarGAN; deep learning; harmonization

Mesh:

Year:  2021        PMID: 34564904      PMCID: PMC8844038          DOI: 10.1002/jmri.27908

Source DB:  PubMed          Journal:  J Magn Reson Imaging        ISSN: 1053-1807            Impact factor:   5.119


  21 in total

1.  On standardizing the MR image intensity scale.

Authors:  L G Nyúl; J K Udupa
Journal:  Magn Reson Med       Date:  1999-12       Impact factor: 4.668

2.  Improved optimization for the robust and accurate linear registration and motion correction of brain images.

Authors:  Mark Jenkinson; Peter Bannister; Michael Brady; Stephen Smith
Journal:  Neuroimage       Date:  2002-10       Impact factor: 6.556

3.  Addressing population aging and Alzheimer's disease through the Australian imaging biomarkers and lifestyle study: collaboration with the Alzheimer's Disease Neuroimaging Initiative.

Authors:  Kathryn A Ellis; Christopher C Rowe; Victor L Villemagne; Ralph N Martins; Colin L Masters; Olivier Salvado; Cassandra Szoeke; David Ames
Journal:  Alzheimers Dement       Date:  2010-05       Impact factor: 21.566

4.  Multi-atlas skull-stripping.

Authors:  Jimit Doshi; Guray Erus; Yangming Ou; Bilwaj Gaonkar; Christos Davatzikos
Journal:  Acad Radiol       Date:  2013-12       Impact factor: 3.173

5.  Evaluating the Impact of Intensity Normalization on MR Image Synthesis.

Authors:  Jacob C Reinhold; Blake E Dewey; Aaron Carass; Jerry L Prince
Journal:  Proc SPIE Int Soc Opt Eng       Date:  2019-03

6.  Evaluating intensity normalization on MRIs of human brain with multiple sclerosis.

Authors:  Mohak Shah; Yiming Xiao; Nagesh Subbanna; Simon Francis; Douglas L Arnold; D Louis Collins; Tal Arbel
Journal:  Med Image Anal       Date:  2010-12-25       Impact factor: 8.545

7.  Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation.

Authors:  Konstantinos Kamnitsas; Christian Ledig; Virginia F J Newcombe; Joanna P Simpson; Andrew D Kane; David K Menon; Daniel Rueckert; Ben Glocker
Journal:  Med Image Anal       Date:  2016-10-29       Impact factor: 8.545

8.  Unsupervised MR harmonization by learning disentangled representations using information bottleneck theory.

Authors:  Lianrui Zuo; Blake E Dewey; Yihao Liu; Yufan He; Scott D Newsome; Ellen M Mowry; Susan M Resnick; Jerry L Prince; Aaron Carass
Journal:  Neuroimage       Date:  2021-09-08       Impact factor: 6.556

9.  Variable generalization performance of a deep learning model to detect pneumonia in chest radiographs: A cross-sectional study.

Authors:  John R Zech; Marcus A Badgeley; Manway Liu; Anthony B Costa; Joseph J Titano; Eric Karl Oermann
Journal:  PLoS Med       Date:  2018-11-06       Impact factor: 11.069

Review 10.  Ten Years of BrainAGE as a Neuroimaging Biomarker of Brain Aging: What Insights Have We Gained?

Authors:  Katja Franke; Christian Gaser
Journal:  Front Neurol       Date:  2019-08-14       Impact factor: 4.003

View more
  1 in total

1.  Applications of Radiomics and Radiogenomics in High-Grade Gliomas in the Era of Precision Medicine.

Authors:  Anahita Fathi Kazerooni; Stephen J Bagley; Hamed Akbari; Sanjay Saxena; Sina Bagheri; Jun Guo; Sanjeev Chawla; Ali Nabavizadeh; Suyash Mohan; Spyridon Bakas; Christos Davatzikos; MacLean P Nasrallah
Journal:  Cancers (Basel)       Date:  2021-11-25       Impact factor: 6.575

  1 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.