Cross-scanner and cross-protocol multi-shell diffusion MRI data harmonization: Algorithms and results
Lipeng Ning,
Elisenda Bonet-Carne,
Francesco Grussu,
Farshid Sepehrband,
Enrico Kaden,
Jelle Veraart,
Stefano B. Blumberg,
Can Son Khoo,
Marco Palombo,
Iasonas Kokkinos,
Daniel C. Alexander,
Jaume Coll-Font,
Benoit Scherrer,
Simon K. Warfield,
Suheyla Cetin Karayumak,
Yogesh Rathi,
Simon Koppers,
Leon Weninger,
Julia Ebert,
Dorit Merhof,
Daniel Moyer,
Maximilian Pietsch,
Daan Christiaens,
Rui Azeredo Gomes Teixeira,
Jacques-Donald Tournier,
Kurt G. Schilling,
Yuankai Huo,
Vishwesh Nath,
Colin Hansen,
Justin Blaber,
Bennett A. Landman,
Andrey Zhylka,
Josien P.W. Pluim,
Greg Parker,
Umesh Rudrapatna,
John Evans,
Cyril Charron,
Derek K. Jones,
Chantal M.W. Tax
Affiliations
Lipeng Ning
Brigham and Women’s Hospital, Boston, United States; Harvard Medical School, Boston, United States; Corresponding author. Brigham and Women’s Hospital, Boston, United States
Elisenda Bonet-Carne
University College London, London, United Kingdom
Francesco Grussu
University College London, London, United Kingdom
Farshid Sepehrband
Stevens Neuroimaging and Informatics Institute, Keck School of Medicine of USC, University of Southern California, Los Angeles, United States
Enrico Kaden
University College London, London, United Kingdom
Jelle Veraart
New York University, New York, NY, United States
Stefano B. Blumberg
University College London, London, United Kingdom
Can Son Khoo
University College London, London, United Kingdom
Marco Palombo
University College London, London, United Kingdom
Iasonas Kokkinos
University College London, London, United Kingdom
Daniel C. Alexander
University College London, London, United Kingdom
Jaume Coll-Font
Boston Children’s Hospital, Boston, United States; Harvard Medical School, Boston, United States
Benoit Scherrer
Boston Children’s Hospital, Boston, United States; Harvard Medical School, Boston, United States
Simon K. Warfield
Boston Children’s Hospital, Boston, United States; Harvard Medical School, Boston, United States
Suheyla Cetin Karayumak
Brigham and Women’s Hospital, Boston, United States; Harvard Medical School, Boston, United States
Yogesh Rathi
Brigham and Women’s Hospital, Boston, United States; Harvard Medical School, Boston, United States
Simon Koppers
RWTH Aachen University, Aachen, Germany
Leon Weninger
RWTH Aachen University, Aachen, Germany
Julia Ebert
RWTH Aachen University, Aachen, Germany
Dorit Merhof
RWTH Aachen University, Aachen, Germany
Daniel Moyer
Stevens Neuroimaging and Informatics Institute, Keck School of Medicine of USC, University of Southern California, Los Angeles, United States
Maximilian Pietsch
Centre for the Developing Brain, School of Biomedical Engineering and Imaging Sciences, King’s College London, London, United Kingdom
Daan Christiaens
Centre for the Developing Brain, School of Biomedical Engineering and Imaging Sciences, King’s College London, London, United Kingdom
Rui Azeredo Gomes Teixeira
Centre for the Developing Brain, School of Biomedical Engineering and Imaging Sciences, King’s College London, London, United Kingdom
Jacques-Donald Tournier
Centre for the Developing Brain, School of Biomedical Engineering and Imaging Sciences, King’s College London, London, United Kingdom
Kurt G. Schilling
Institute of Imaging Science, Vanderbilt University, Nashville, TN, United States
Yuankai Huo
Department of Electrical Engineering & Computer Science, Vanderbilt University, Nashville, TN, United States
Vishwesh Nath
Department of Electrical Engineering & Computer Science, Vanderbilt University, Nashville, TN, United States
Colin Hansen
Department of Electrical Engineering & Computer Science, Vanderbilt University, Nashville, TN, United States
Justin Blaber
Department of Electrical Engineering & Computer Science, Vanderbilt University, Nashville, TN, United States
Bennett A. Landman
Institute of Imaging Science, Vanderbilt University, Nashville, TN, United States; Department of Electrical Engineering & Computer Science, Vanderbilt University, Nashville, TN, United States; Department of Biomedical Engineering, Vanderbilt University, Nashville, TN, United States
Andrey Zhylka
Eindhoven University of Technology, Eindhoven, Netherlands
Josien P.W. Pluim
Eindhoven University of Technology, Eindhoven, Netherlands
Greg Parker
Cardiff University Brain Research Imaging Centre (CUBRIC), Cardiff University, Cardiff, United Kingdom
Umesh Rudrapatna
Cardiff University Brain Research Imaging Centre (CUBRIC), Cardiff University, Cardiff, United Kingdom
John Evans
Cardiff University Brain Research Imaging Centre (CUBRIC), Cardiff University, Cardiff, United Kingdom
Cyril Charron
Cardiff University Brain Research Imaging Centre (CUBRIC), Cardiff University, Cardiff, United Kingdom
Derek K. Jones
Cardiff University Brain Research Imaging Centre (CUBRIC), Cardiff University, Cardiff, United Kingdom; School of Psychology, Australian Catholic University, Melbourne, Australia
Chantal M.W. Tax
Cardiff University Brain Research Imaging Centre (CUBRIC), Cardiff University, Cardiff, United Kingdom
Cross-scanner and cross-protocol variability of diffusion magnetic resonance imaging (dMRI) data are known to be major obstacles in multi-site clinical studies since they limit the ability to aggregate dMRI data and derived measures. Computational algorithms that harmonize the data and minimize such variability are critical to reliably combine datasets acquired from different scanners and/or protocols, thus improving the statistical power and sensitivity of multi-site studies. Different computational approaches have been proposed to harmonize diffusion MRI data or remove scanner-specific differences. To date, these methods have mostly been developed for or evaluated on single b-value diffusion MRI data. In this work, we present the evaluation results of 19 algorithms that are developed to harmonize the cross-scanner and cross-protocol variability of multi-shell diffusion MRI using a benchmark database. The proposed algorithms rely on various signal representation approaches and computational tools, such as rotational invariant spherical harmonics, deep neural networks and hybrid biophysical and statistical approaches. The benchmark database consists of data acquired from the same subjects on two scanners with different maximum gradient strength (80 and 300 mT/m) and with two protocols. We evaluated the performance of these algorithms for mapping multi-shell diffusion MRI data across scanners and across protocols using several state-of-the-art imaging measures. The results show that data harmonization algorithms can reduce the cross-scanner and cross-protocol variabilities to a similar level as scan-rescan variability using the same scanner and protocol. In particular, the LinearRISH algorithm based on adaptive linear mapping of rotational invariant spherical harmonics features yields the lowest variability for our data in predicting the fractional anisotropy (FA), mean diffusivity (MD), mean kurtosis (MK) and the rotationally invariant spherical harmonic (RISH) features. But other algorithms, such as DIAMOND, SHResNet, DIQT, CMResNet show further improvement in harmonizing the return-to-origin probability (RTOP). The performance of different approaches provides useful guidelines on data harmonization in future multi-site studies.