TY - JOUR
T1 - Accuracy and reliability of diffusion imaging models
AU - Seider, Nicole A.
AU - Adeyemo, Babatunde
AU - Miller, Ryland
AU - Newbold, Dillan J.
AU - Hampton, Jacqueline M.
AU - Scheidter, Kristen M.
AU - Rutlin, Jerrel
AU - Laumann, Timothy O.
AU - Roland, Jarod L.
AU - Montez, David F.
AU - Van, Andrew N.
AU - Zheng, Annie
AU - Marek, Scott
AU - Kay, Benjamin P.
AU - Bretthorst, G.
AU - Schlaggar, Bradley L.
AU - Greene, Deanna
AU - Wang, Yong
AU - Petersen, Steven E.
AU - Barch, Deanna M.
AU - Gordon, Evan M.
AU - Snyder, Abraham Z.
AU - Shimony, Joshua S.
AU - Dosenbach, Nico U.F.
N1 - Funding Information:
This work was supported by NIH grants T32MH100019 (N.A.S), NS088590 (N.U.F.D.), MH96773 (N.U.F.D.), MH122066 (N.U.F.D.), MH124567 (N.U.F.D.), MH121276 (N.U.F.D.), HD087011 (J.S.S.), NS115672 (A.Z.), NS110332 (D.J.N.), MH1000872 (T.O.L.), MH112473 (T.O.L.), NS090978 (B.P.K.), MH121518 (S.M.), MH104592 (D.J.G.), HD094381 (Y.W.), AG053548 (Y.W.), NS098577 (to the Neuroimaging Informatics and Analysis Center); the US Department of Veterans Affairs Clinical Sciences Research and Development Service grant 1IK2CX001680 (E.M.G.); Kiwanis Neuroscience Research Foundation (N.U.F.D.); the Jacobs Foundation grant 2016121703 (N.U.F.D.); the Child Neurology Foundation (N.U.F.D.); the McDonnell Center for Systems Neuroscience (D.J.N., T.O.L., A.Z., B.L.S., and N.U.F.D.); the McDonnell Foundation (S.E.P.), the Mallinckrodt Institute of Radiology grant 14–011 (N.U.F.D.); the Hope Center for Neurological Disorders (N.U.F.D., B.L.S., and S.E.P.); the Intellectual and Developmental Disabilities Research Center at Washington University (J.S.S); the BrightFocus Foundation grant A2017330S (Y.W.); the March of Dimes (Y.W.).
Publisher Copyright:
© 2022
PY - 2022/7/1
Y1 - 2022/7/1
N2 - Diffusion imaging aims to non-invasively characterize the anatomy and integrity of the brain's white matter fibers. We evaluated the accuracy and reliability of commonly used diffusion imaging methods as a function of data quantity and analysis method, using both simulations and highly sampled individual-specific data (927–1442 diffusion weighted images [DWIs] per individual). Diffusion imaging methods that allow for crossing fibers (FSL's BedpostX [BPX], DSI Studio's Constant Solid Angle Q-Ball Imaging [CSA-QBI], MRtrix3's Constrained Spherical Deconvolution [CSD]) estimated excess fibers when insufficient data were present and/or when the data did not match the model priors. To reduce such overfitting, we developed a novel Bayesian Multi-tensor Model-selection (BaMM) method and applied it to the popular ball-and-stick model used in BedpostX within the FSL software package. BaMM was robust to overfitting and showed high reliability and the relatively best crossing-fiber accuracy with increasing amounts of diffusion data. Thus, sufficient data and an overfitting resistant analysis method enhance precision diffusion imaging. For potential clinical applications of diffusion imaging, such as neurosurgical planning and deep brain stimulation (DBS), the quantities of data required to achieve diffusion imaging reliability are lower than those needed for functional MRI.
AB - Diffusion imaging aims to non-invasively characterize the anatomy and integrity of the brain's white matter fibers. We evaluated the accuracy and reliability of commonly used diffusion imaging methods as a function of data quantity and analysis method, using both simulations and highly sampled individual-specific data (927–1442 diffusion weighted images [DWIs] per individual). Diffusion imaging methods that allow for crossing fibers (FSL's BedpostX [BPX], DSI Studio's Constant Solid Angle Q-Ball Imaging [CSA-QBI], MRtrix3's Constrained Spherical Deconvolution [CSD]) estimated excess fibers when insufficient data were present and/or when the data did not match the model priors. To reduce such overfitting, we developed a novel Bayesian Multi-tensor Model-selection (BaMM) method and applied it to the popular ball-and-stick model used in BedpostX within the FSL software package. BaMM was robust to overfitting and showed high reliability and the relatively best crossing-fiber accuracy with increasing amounts of diffusion data. Thus, sufficient data and an overfitting resistant analysis method enhance precision diffusion imaging. For potential clinical applications of diffusion imaging, such as neurosurgical planning and deep brain stimulation (DBS), the quantities of data required to achieve diffusion imaging reliability are lower than those needed for functional MRI.
UR - http://www.scopus.com/inward/record.url?scp=85127097041&partnerID=8YFLogxK
U2 - 10.1016/j.neuroimage.2022.119138
DO - 10.1016/j.neuroimage.2022.119138
M3 - Article
C2 - 35339687
AN - SCOPUS:85127097041
SN - 1053-8119
VL - 254
JO - NeuroImage
JF - NeuroImage
M1 - 119138
ER -