PMID- 33722670 OWN - NLM STAT- MEDLINE DCOM- 20211013 LR - 20211013 IS - 1095-9572 (Electronic) IS - 1053-8119 (Linking) VI - 234 DP - 2021 Jul 1 TI - Decoding with confidence: Statistical control on decoder maps. PG - 117921 LID - S1053-8119(21)00198-1 [pii] LID - 10.1016/j.neuroimage.2021.117921 [doi] AB - In brain imaging, decoding is widely used to infer relationships between brain and cognition, or to craft brain-imaging biomarkers of pathologies. Yet, standard decoding procedures do not come with statistical guarantees, and thus do not give confidence bounds to interpret the pattern maps that they produce. Indeed, in whole-brain decoding settings, the number of explanatory variables is much greater than the number of samples, hence classical statistical inference methodology cannot be applied. Specifically, the standard practice that consists in thresholding decoding maps is not a correct inference procedure. We contribute a new statistical-testing framework for this type of inference. To overcome the statistical inefficiency of voxel-level control, we generalize the Family Wise Error Rate (FWER) to account for a spatial tolerance delta, introducing the delta-Family Wise Error Rate (delta-FWER). Then, we present a decoding procedure that can control the delta-FWER: the Ensemble of Clustered Desparsified Lasso (EnCluDL), a procedure for multivariate statistical inference on high-dimensional structured data. We evaluate the statistical properties of EnCluDL with a thorough empirical study, along with three alternative procedures including decoder map thresholding. We show that EnCluDL exhibits the best recovery properties while ensuring the expected statistical control. CI - Copyright (c) 2021 The Authors. Published by Elsevier Inc. All rights reserved. FAU - Chevalier, Jerome-Alexis AU - Chevalier JA AD - Parietal project-team, Inria Saclay-Ile de France, Palaiseau, France; CEA/Neurospin bat 145, Gif-Sur-Yvette, France; Universite Paris-Saclay, Gif-Sur-Yvette, France. Electronic address: jerome-alexis.chevalier@inria.fr. FAU - Nguyen, Tuan-Binh AU - Nguyen TB AD - Parietal project-team, Inria Saclay-Ile de France, Palaiseau, France; CEA/Neurospin bat 145, Gif-Sur-Yvette, France; Universite Paris-Saclay, Gif-Sur-Yvette, France; LMO, Universite Paris-Saclay, Orsay, France. FAU - Salmon, Joseph AU - Salmon J AD - IMAG, Universite de Montpellier, CNRS, Montpellier, France. FAU - Varoquaux, Gael AU - Varoquaux G AD - Parietal project-team, Inria Saclay-Ile de France, Palaiseau, France; CEA/Neurospin bat 145, Gif-Sur-Yvette, France; Universite Paris-Saclay, Gif-Sur-Yvette, France. FAU - Thirion, Bertrand AU - Thirion B AD - Parietal project-team, Inria Saclay-Ile de France, Palaiseau, France; CEA/Neurospin bat 145, Gif-Sur-Yvette, France; Universite Paris-Saclay, Gif-Sur-Yvette, France. LA - eng GR - U54 MH091657/MH/NIMH NIH HHS/United States PT - Journal Article PT - Research Support, N.I.H., Extramural PT - Research Support, Non-U.S. Gov't DEP - 20210312 PL - United States TA - Neuroimage JT - NeuroImage JID - 9215515 SB - IM MH - Algorithms MH - Brain/*diagnostic imaging/*physiology MH - Brain Mapping/*methods MH - *Data Analysis MH - Humans MH - Image Processing, Computer-Assisted/methods MH - Magnetic Resonance Imaging/*methods MH - *Neural Networks, Computer MH - Psychomotor Performance/physiology OTO - NOTNLM OT - Decoding OT - High dimension OT - Inference OT - Multivariate model OT - Statistical control OT - Statistical methods OT - Support recovery OT - fMRI EDAT- 2021/03/17 06:00 MHDA- 2021/10/14 06:00 CRDT- 2021/03/16 06:09 PHST- 2020/04/30 00:00 [received] PHST- 2021/02/17 00:00 [revised] PHST- 2021/02/21 00:00 [accepted] PHST- 2021/03/17 06:00 [pubmed] PHST- 2021/10/14 06:00 [medline] PHST- 2021/03/16 06:09 [entrez] AID - S1053-8119(21)00198-1 [pii] AID - 10.1016/j.neuroimage.2021.117921 [doi] PST - ppublish SO - Neuroimage. 2021 Jul 1;234:117921. doi: 10.1016/j.neuroimage.2021.117921. Epub 2021 Mar 12.