Skip to Main content Skip to Navigation
Journal articles

Power estimation for non-standardized multisite studies

Anisha Keshavan 1, 2 Friedemann Paul 3, 4 Mona K. Beyer 5 Alyssa H. Zhu 2 Nico Papinutto 2 Russell T. Shinohara 6 William Stern 2 Michael Amann 7, 8 Rohit Bakshi 9 Antje Bischof 10 Alessandro Carriero 11 Manuel Comabella 12 Jason C. Crane 13 Sandra d'Alfonso 14 Philippe Demaerel 15 Benedicte Dubois 16 Massimo Filippi 17 Vinzenz Fleischer 18, 19 Bertrand Fontaine 20 Laura Gaetano 7, 8 An Goris 16 Christiane Graetz 19 Adriane Gröger 19 Sergiu Groppa 19 David A. Hafler 21 Hanne F. Harbo 22 Bernhard Hemmer 23, 24 Kesshi Jordan 2, 1 Ludwig Kappos 7 Gina Kirkish 13 Sara Llufriu 25 Stefano Magon 7, 8 Filippo Martinelli-Boneschi 26 Jacob L. Mccauley 27 Xavier Montalban 8, 12 Mark Mühlau 23, 28 Daniel Pelletier 21 Pradip M. Pattany 29 Margaret Pericak-Vance 27 Isabelle Cournu-Rebeix 20 Maria A. Rocca 26 Alex Rovira 12 Regina Schlaeger 2, 7, 10 Albert Saiz 27 Till Sprenger 30 Alessandro Stecco 31 Bernard M.J. Uitdehaag 32 Pablo Villoslada 2, 25 Mike P. Wattjes 32 Howard Weiner 9 Jens Wuerfel 4, 8 Claus Zimmer 33 Frauke Zipp 19 Stephen L. Hauser 2 Jorge R. Oksenberg 2 Roland G. Henry 1, 2, 13
Abstract : A concern for researchers planning multisite studies is that scanner and T1-weighted sequence-related biases on regional volumes could overshadow true effects, especially for studies with a heterogeneous set of scanners and sequences. Current approaches attempt to harmonize data by standardizing hardware, pulse sequences, and protocols, or by calibrating across sites using phantom-based corrections to ensure the same raw image intensities. We propose to avoid harmonization and phantom-based correction entirely. We hypothesized that the bias of estimated regional volumes is scaled between sites due to the contrast and gradient distortion differences between scanners and sequences. Given this assumption, we provide a new statistical framework and derive a power equation to define inclusion criteria for a set of sites based on the variability of their scaling factors. We estimated the scaling factors of 20 scanners with heterogeneous hardware and sequence parameters by scanning a single set of 12 subjects at sites across the United States and Europe. Regional volumes and their scaling factors were estimated for each site using Freesurfer's segmentation algorithm and ordinary least squares, respectively. The scaling factors were validated by comparing the theoretical and simulated power curves, performing a leave-one-out calibration of regional volumes, and evaluating the absolute agreement of all regional volumes between sites before and after calibration. Using our derived power equation, we were able to define the conditions under which harmonization is not necessary to achieve 80% power. This approach can inform choice of processing pipelines and outcome metrics for multisite studies based on scaling factor variability across sites, enabling collaboration between clinical and research institutions.
Document type :
Journal articles
Complete list of metadatas

Cited literature [54 references]  Display  Hide  Download
Contributor : Gilles d'Eggis <>
Submitted on : Wednesday, May 4, 2016 - 2:17:18 PM
Last modification on : Thursday, August 22, 2019 - 11:02:05 AM
Document(s) archivé(s) le : Tuesday, November 15, 2016 - 8:15:42 PM


Publication funded by an institution


Distributed under a Creative Commons Attribution 4.0 International License



Anisha Keshavan, Friedemann Paul, Mona K. Beyer, Alyssa H. Zhu, Nico Papinutto, et al.. Power estimation for non-standardized multisite studies. NeuroImage, Elsevier, 2016, 134, ⟨10.1016/j.neuroimage.2016.03.051⟩. ⟨hal-01311572⟩



Record views


Files downloads