Academic Journal
View in EDS

Peaks2Image: Reconstructing fMRI maps from stereotactic coordinates to enhance cognitive meta-analysis.

Bibliographic Details
Title: Peaks2Image: Reconstructing fMRI maps from stereotactic coordinates to enhance cognitive meta-analysis.
Authors: Meudec R; Université Paris-Saclay, Inria, CEA, Palaiseau, France., Dockès J; Université Paris-Saclay, Inria, CEA, Palaiseau, France., Ghayem F; Université Paris-Saclay, Inria, CEA, Palaiseau, France., Wassermann D; Université Paris-Saclay, Inria, CEA, Palaiseau, France., Thirion B; Université Paris-Saclay, Inria, CEA, Palaiseau, France.
Source: Imaging neuroscience (Cambridge, Mass.) [Imaging Neurosci (Camb)] 2026 Sep 18; Vol. 4. Date of Electronic Publication: 2026 Sep 18 (Print Publication: 2026).
Publication Type: Journal Article
Language: English
Journal Info: Publisher: The MIT Press Country of Publication: United States NLM ID: 9918663686606676 Publication Model: eCollection Cited Medium: Internet ISSN: 2837-6056 (Electronic) Linking ISSN: 28376056 NLM ISO Abbreviation: Imaging Neurosci (Camb) Subsets: MEDLINE
Imprint Name(s): Original Publication: Cambridge, Massachusetts : The MIT Press, [2023]-
MeSH Terms: Magnetic Resonance Imaging*/methods , Brain Mapping*/methods , Brain*/diagnostic imaging , Brain*/physiology , Image Processing, Computer-Assisted*/methods , Cognition*/physiology , Meta-Analysis as Topic*, Humans ; Neural Networks, Computer ; Stereotaxic Techniques
Abstract: To integrate novel findings and reach a consensus on the neural correlates of cognitive functions, human neuroscientists have relied heavily on the literature to formulate, validate, or refine hypotheses. Since thousands of publications have reported the coordinates of regions identified in brain mapping, the community has largely performed meta-analyses based on these coordinates. A known problem with coordinate-based meta-analyses is that they rely on limited information from each publication, so terms appearing in few publications are not reliably mapped. Recently, the use of public images has stimulated meta-analysis, which has become possible due to the creation of large image repositories. NeuroVault is a prominent example of such a repository. These repositories allow researchers to compare images across publications, leading to image-based meta-analyses. However, image-based meta-analysis faces a major obstacle: the scarcity and inconsistency of image annotation and metadata. In this paper, we propose Peaks2Image, a neural network approach that reconstructs continuous spatial representations of brain activity from peak activation tables. Peaks2Image provides dense brain reconstructions that can be combined with extensive annotations from the neuroscience literature. We use these annotations to train a decoder with textual features as labels. This results in a much broader set of decoded terms than current image-based studies. We validated the decoder using 47,000 annotated images and successfully identified 91 out of 108 terms, with a model that was uniquely trained using the neuroscience literature.
(© 2026 The Authors. Published under a Creative Commons Attribution 4.0 International (CC BY 4.0) license.)
Competing Interests: The authors declare no competing interests.
References: Soc Cogn Affect Neurosci. 2007 Jun;2(2):150-8. (PMID: 18985131)
Neuroimage. 2003 Jun;19(2 Pt 1):261-70. (PMID: 12814577)
Neuroimage. 2018 Oct 15;180(Pt A):243-252. (PMID: 29448074)
Neuroinformatics. 2005;3(1):65-78. (PMID: 15897617)
Neuroimage. 2022 Nov;263:119610. (PMID: 36064138)
Neuroimage. 2006 Aug 15;32(2):551-7. (PMID: 16815038)
Annu Rev Psychol. 2016;67:587-612. (PMID: 26393866)
Neuroimage. 2013 Oct 15;80:144-68. (PMID: 23702415)
Gigascience. 2022 Dec 28;12:. (PMID: 37132522)
Nat Methods. 2011 Jun 26;8(8):665-70. (PMID: 21706013)
PLoS Comput Biol. 2019 May 24;15(5):e1006299. (PMID: 31125335)
J Am Med Inform Assoc. 2024 Sep 1;31(9):1833-1843. (PMID: 38613821)
Front Neuroinform. 2011 Sep 06;5:17. (PMID: 21922006)
Front Neuroinform. 2015 Apr 10;9:8. (PMID: 25914639)
Nat Neurosci. 2021 Dec;24(12):1733-1744. (PMID: 34764476)
Nat Rev Neurosci. 2017 Feb;18(2):115-126. (PMID: 28053326)
Neuroimage. 2019 Jan 1;184:335-348. (PMID: 30237036)
Proc Natl Acad Sci U S A. 2016 Nov 1;113(44):12574-12579. (PMID: 27791099)
Neuron. 2011 Dec 8;72(5):692-7. (PMID: 22153367)
Elife. 2025 Sep 11;13:. (PMID: 40934113)
Neuroimage. 2009 Apr 15;45(3):810-23. (PMID: 19166944)
Imaging Neurosci (Camb). 2026 Mar 16;4:. (PMID: 41852941)
Sci Rep. 2022 Apr 29;12(1):7050. (PMID: 35488032)
Elife. 2020 Mar 04;9:. (PMID: 32129761)
Curr Opin Neurobiol. 2019 Apr;55:1-6. (PMID: 30513462)
PLoS Comput Biol. 2018 Nov 29;14(11):e1006565. (PMID: 30496171)
Netw Neurosci. 2023 Jun 30;7(2):769-786. (PMID: 37397893)
Front Hum Neurosci. 2025 Jul 10;19:1580808. (PMID: 40708807)
Neuroimage. 2020 Nov 1;221:117126. (PMID: 32673748)
Contributed Indexing: Keywords: DiFuMo; coordinate-based meta-analysis; decoding; fMRI; image reconstruction; set transformer
Entry Date(s): Date Created: 20260922 Date Completed: 20260922 Latest Revision: 20260924
Update Code: 20260924
PubMed Central ID: PMC13591505
DOI: 10.1162/IMAG.a.1367
PMID: 42769493
Database: MEDLINE
Description
ISSN:2837-6056
DOI:10.1162/IMAG.a.1367