{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T20:12:46Z","timestamp":1789762366440,"version":"4.0.1"},"reference-count":46,"publisher":"MIT Press","license":[{"start":{"date-parts":[[2026,8,26]],"date-time":"2026-08-26T00:00:00Z","timestamp":1787702400000},"content-version":"vor","delay-in-days":237,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["direct.mit.edu"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,9,18]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>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.<\/jats:p>","DOI":"10.1162\/imag.a.1367","type":"journal-article","created":{"date-parts":[[2026,8,26]],"date-time":"2026-08-26T20:01:51Z","timestamp":1787774511000},"update-policy":"https:\/\/doi.org\/10.1162\/mitpressjournals.corrections.policy","source":"Crossref","is-referenced-by-count":0,"title":["Peaks2Image: Reconstructing fMRI maps from stereotactic coordinates to enhance cognitive meta-analysis"],"prefix":"10.1162","volume":"4","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-9970-5745","authenticated-orcid":false,"given":"Rapha\u00ebl","family":"Meudec","sequence":"first","affiliation":[{"name":"Universit\u00e9 Paris-Saclay, Inria, CEA, Palaiseau, France"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"J\u00e9r\u00f4me","family":"Dock\u00e8s","sequence":"additional","affiliation":[{"name":"Universit\u00e9 Paris-Saclay, Inria, CEA, Palaiseau, France"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fateme","family":"Ghayem","sequence":"additional","affiliation":[{"name":"Universit\u00e9 Paris-Saclay, Inria, CEA, Palaiseau, France"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Demian","family":"Wassermann","sequence":"additional","affiliation":[{"name":"Universit\u00e9 Paris-Saclay, Inria, CEA, Palaiseau, France"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5018-7895","authenticated-orcid":false,"given":"Bertrand","family":"Thirion","sequence":"additional","affiliation":[{"name":"Universit\u00e9 Paris-Saclay, Inria, CEA, Palaiseau, France"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"281","published-online":{"date-parts":[[2026,9,18]]},"reference":[{"issue":"12","key":"2026091815555570700_IMAG.a.1367-b1","doi-asserted-by":"crossref","first-page":"1733","DOI":"10.1038\/s41593-021-00948-9","article-title":"A data-driven framework for mapping domains of human neurobiology","volume":"24","author":"Beam","year":"2021","journal-title":"Nature Neuroscience"},{"key":"2026091815555570700_IMAG.a.1367-b2","doi-asserted-by":"crossref","unstructured":"Beltagy, I., Lo, K., & Cohan, A. (2019). SciBERT: A pretrained language model for scientific text. arXiv. https:\/\/doi.org\/10.48550\/arXiv.1903.10676","DOI":"10.18653\/v1\/D19-1371"},{"key":"2026091815555570700_IMAG.a.1367-b3","first-page":"29","article-title":"A Bayesian method for reducing bias in neural representational similarity analysis","author":"Cai","year":"2016","journal-title":"Advances in Neural Information Processing Systems"},{"key":"2026091815555570700_IMAG.a.1367-b4","unstructured":"Careil, M., Benchetrit, Y., & King, J.-R. (2025). Dynadiff: Single-stage decoding of images from continuously evolving fMRI. arXiv. https:\/\/doi.org\/10.48550\/arXiv.2505.14556"},{"key":"2026091815555570700_IMAG.a.1367-b5","doi-asserted-by":"crossref","unstructured":"Caro, J. O., de Oliveira Fonseca, A. H., Rizvi, S. A., Rosati, M., Averill, C. L., Cross, J. L., Mittal, P., Zappala, E., Dhodapkar, R. M., Abdallah, C., & van Dijk, D. (2024). BrainLM: A foundation model for brain activity recordings. bioRxiv. https:\/\/doi.org\/10.1101\/2023.09.12.557460","DOI":"10.1101\/2023.09.12.557460"},{"issue":"2","key":"2026091815555570700_IMAG.a.1367-b6","doi-asserted-by":"crossref","first-page":"261","DOI":"10.1016\/S1053-8119(03)00049-1","article-title":"Functional magnetic resonance imaging (fMRI) \u201cbrain reading\u201d: Detecting and classifying distributed patterns of fMRI activity in human visual cortex","volume":"19","author":"Cox","year":"2003","journal-title":"NeuroImage"},{"key":"2026091815555570700_IMAG.a.1367-b7","doi-asserted-by":"crossref","first-page":"117126","DOI":"10.1016\/j.neuroimage.2020.117126","article-title":"Fine-grain atlases of functional modes for fMRI analysis","volume":"221","author":"Dadi","year":"2020","journal-title":"NeuroImage"},{"key":"2026091815555570700_IMAG.a.1367-b8","doi-asserted-by":"crossref","first-page":"335","DOI":"10.1016\/j.neuroimage.2018.09.042","article-title":"Distinct modes of functional connectivity induced by movie-watching","volume":"184","author":"Demirta\u015f","year":"2019","journal-title":"NeuroImage"},{"issue":"3","key":"2026091815555570700_IMAG.a.1367-b9","doi-asserted-by":"crossref","first-page":"297","DOI":"10.2307\/1932409","article-title":"Measures of the amount of ecologic association between species","volume":"26","author":"Dice","year":"1945","journal-title":"Ecology"},{"key":"2026091815555570700_IMAG.a.1367-b10","doi-asserted-by":"crossref","unstructured":"Dock\u00e8s, J., Oudyk, K., Torabi, M., de la Vega, A. I., & Poline, J.-B. (2023). Mining the neuroimaging literature. bioRxiv. https:\/\/doi.org\/10.7554\/elife.94909","DOI":"10.1101\/2023.10.30.564783"},{"key":"2026091815555570700_IMAG.a.1367-b11","doi-asserted-by":"crossref","first-page":"e53385","DOI":"10.7554\/eLife.53385","article-title":"NeuroQuery, comprehensive meta-analysis of human brain mapping","volume":"9","author":"Dock\u00e8s","year":"2020","journal-title":"Elife"},{"key":"2026091815555570700_IMAG.a.1367-b12","article-title":"On the benefits of self-taught learning for brain decoding","volume":"12","author":"Germani","year":"2023","journal-title":"GigaScience"},{"key":"2026091815555570700_IMAG.a.1367-b13","article-title":"NeuroConText: Contrastive learning for neuroscience meta-analysis with rich text representation","volume":"4","author":"Ghayem","year":"2026","journal-title":"Imaging Neuroscience (Camb.)"},{"key":"2026091815555570700_IMAG.a.1367-b14","article-title":"OpenNeuro\u2014A free online platform for sharing and analysis of neuroimaging data","volume":"1677","author":"Gorgolewski","year":"2017","journal-title":"Organization for Human Brain Mapping, Vancouver, Canada"},{"key":"2026091815555570700_IMAG.a.1367-b15","doi-asserted-by":"crossref","first-page":"8","DOI":"10.3389\/fninf.2015.00008","article-title":"Neurovault.org: A web-based repository for collecting and sharing unthresholded statistical maps of the human brain","volume":"9","author":"Gorgolewski","year":"2015","journal-title":"Frontiers in Neuroinformatics"},{"key":"2026091815555570700_IMAG.a.1367-b16","first-page":"57","article-title":"peaks2maps: Reconstructing unthresholded statistical maps from peak coordinates using deep neural networks","volume":"8","author":"Gorgolewski","year":"2019","journal-title":"F1000Research"},{"key":"2026091815555570700_IMAG.a.1367-b17","article-title":"Image-to-image translation with conditional adversarial networks","author":"Isola","year":"2016","journal-title":"arXiv"},{"issue":"1","key":"2026091815555570700_IMAG.a.1367-b18","doi-asserted-by":"crossref","first-page":"65","DOI":"10.1385\/NI:3:1:065","article-title":"BrainMap: The social evolution of a human brain mapping database","volume":"3","author":"Laird","year":"2005","journal-title":"Neuroinformatics"},{"key":"2026091815555570700_IMAG.a.1367-b19","article-title":"Set transformer: A framework for attention-based permutation-invariant neural networks","author":"Lee","year":"2019","journal-title":"arXiv"},{"key":"2026091815555570700_IMAG.a.1367-b20","doi-asserted-by":"crossref","first-page":"243","DOI":"10.1016\/j.neuroimage.2018.01.071","article-title":"A probabilistic approach to discovering dynamic full-brain functional connectivity patterns","volume":"180","author":"Manning","year":"2018","journal-title":"NeuroImage"},{"issue":"44","key":"2026091815555570700_IMAG.a.1367-b21","doi-asserted-by":"crossref","first-page":"12574","DOI":"10.1073\/pnas.1608282113","article-title":"Situating the default-mode network along a principal gradient of macroscale cortical organization","volume":"113","author":"Margulies","year":"2016","journal-title":"Proceedings of the National Academy of Sciences of the United States of America"},{"key":"2026091815555570700_IMAG.a.1367-b22","article-title":"Learning neural representations of human cognition across many fMRI studies","volume":"30","author":"Mensch","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"issue":"1","key":"2026091815555570700_IMAG.a.1367-b23","doi-asserted-by":"crossref","first-page":"7050","DOI":"10.1038\/s41598-022-10710-1","article-title":"Comprehensive decoding mental processes from web repositories of functional brain images","volume":"12","author":"Menuet","year":"2022","journal-title":"Scientific Reports"},{"key":"2026091815555570700_IMAG.a.1367-b24","first-page":"605","article-title":"Text2Brain: Synthesis of brain activation maps from free-form text query","volume-title":"International Conference on Medical Image Computing and Computer-Assisted Intervention","author":"Ngo","year":"2021"},{"key":"2026091815555570700_IMAG.a.1367-b25","doi-asserted-by":"crossref","unstructured":"Pinaya, W. H. L., Tudosiu, P.-D., Dafflon, J., da Costa, P. F., Fernandez, V., Nachev, P., Ourselin, S., & Cardoso, M. J. (2022). Brain imaging generation with latent diffusion models. arXiv. https:\/\/doi.org\/10.48550\/arXiv.2209.07162","DOI":"10.1007\/978-3-031-18576-2_12"},{"issue":"5","key":"2026091815555570700_IMAG.a.1367-b26","doi-asserted-by":"crossref","first-page":"692","DOI":"10.1016\/j.neuron.2011.11.001","article-title":"Inferring mental states from neuroimaging data: From reverse inference to large-scale decoding","volume":"72","author":"Poldrack","year":"2011","journal-title":"Neuron"},{"issue":"2","key":"2026091815555570700_IMAG.a.1367-b27","doi-asserted-by":"crossref","first-page":"115","DOI":"10.1038\/nrn.2016.167","article-title":"Scanning the horizon: Towards transparent and reproducible neuroimaging research","volume":"18","author":"Poldrack","year":"2017","journal-title":"Nature Reviews Neuroscience"},{"key":"2026091815555570700_IMAG.a.1367-b28","doi-asserted-by":"crossref","first-page":"17","DOI":"10.3389\/fninf.2011.00017","article-title":"The cognitive atlas: Toward a knowledge foundation for cognitive neuroscience","volume":"5","author":"Poldrack","year":"2011","journal-title":"Frontiers in Neuroinformatics"},{"key":"2026091815555570700_IMAG.a.1367-b29","doi-asserted-by":"crossref","first-page":"587","DOI":"10.1146\/annurev-psych-122414-033729","article-title":"From brain maps to cognitive ontologies: Informatics and the search for mental structure","volume":"67","author":"Poldrack","year":"2016","journal-title":"Annual Review of Psychology"},{"key":"2026091815555570700_IMAG.a.1367-b30","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"Proceedings of the 38th International Conference on Machine Learning","author":"Radford","year":"2021"},{"issue":"3","key":"2026091815555570700_IMAG.a.1367-b31","doi-asserted-by":"crossref","first-page":"810","DOI":"10.1016\/j.neuroimage.2008.12.039","article-title":"Meta-analysis of neuroimaging data: A comparison of image-based and coordinate-based pooling of studies","volume":"45","author":"Salimi-Khorshidi","year":"2009","journal-title":"NeuroImage"},{"issue":"5","key":"2026091815555570700_IMAG.a.1367-b32","doi-asserted-by":"crossref","first-page":"513","DOI":"10.1016\/0306-4573(88)90021-0","article-title":"Term-weighting approaches in automatic text retrieval","volume":"24","author":"Salton","year":"1988","journal-title":"Information Processing & Management"},{"key":"2026091815555570700_IMAG.a.1367-b33","doi-asserted-by":"crossref","first-page":"144","DOI":"10.1016\/j.neuroimage.2013.05.039","article-title":"Resting-state fMRI in the Human Connectome Project","volume":"80","author":"Smith","year":"2013","journal-title":"NeuroImage"},{"key":"2026091815555570700_IMAG.a.1367-b35","doi-asserted-by":"crossref","DOI":"10.52202\/068431-1545","article-title":"Self-supervised learning of brain dynamics from broad neuroimaging data","volume-title":"Proceedings of the 36th International Conference on Neural Information Processing Systems","author":"Thomas","year":"2022"},{"key":"2026091815555570700_IMAG.a.1367-b36","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023","journal-title":"arXiv"},{"issue":"2","key":"2026091815555570700_IMAG.a.1367-b37","doi-asserted-by":"crossref","first-page":"551","DOI":"10.1016\/j.neuroimage.2006.04.182","article-title":"Effect of spatial smoothing on physiological noise in high-resolution fMRI","volume":"32","author":"Triantafyllou","year":"2006","journal-title":"NeuroImage"},{"issue":"2","key":"2026091815555570700_IMAG.a.1367-b38","doi-asserted-by":"crossref","first-page":"769","DOI":"10.1162\/netn_a_00302","article-title":"An information-theoretic analysis of resting-state versus task fMRI","volume":"7","author":"Tuominen","year":"2023","journal-title":"Network Neuroscience"},{"key":"2026091815555570700_IMAG.a.1367-b39","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.conb.2018.11.002","article-title":"Predictive models avoid excessive reductionism in cognitive neuroimaging","volume":"55","author":"Varoquaux","year":"2019","journal-title":"Current Opinion in Neurobiology"},{"issue":"11","key":"2026091815555570700_IMAG.a.1367-b40","doi-asserted-by":"crossref","first-page":"e1006565","DOI":"10.1371\/journal.pcbi.1006565","article-title":"Atlases of cognition with large-scale human brain mapping","volume":"14","author":"Varoquaux","year":"2018","journal-title":"PLoS Computational Biology"},{"issue":"2","key":"2026091815555570700_IMAG.a.1367-b41","doi-asserted-by":"crossref","first-page":"150","DOI":"10.1093\/scan\/nsm015","article-title":"Meta-analysis of functional neuroimaging data: Current and future directions","volume":"2","author":"Wager","year":"2007","journal-title":"Social Cognitive and Affective Neuroscience"},{"key":"2026091815555570700_IMAG.a.1367-b42","doi-asserted-by":"crossref","first-page":"119610","DOI":"10.1016\/j.neuroimage.2022.119610","article-title":"Predicting brain activation maps for arbitrary tasks with cognitive encoding models","volume":"263","author":"Walters","year":"2022","journal-title":"NeuroImage"},{"key":"2026091815555570700_IMAG.a.1367-b43","article-title":"PMC-LLaMA: Towards building open-source language models for medicine","author":"Wu","year":"2023","journal-title":"arXiv"},{"issue":"8","key":"2026091815555570700_IMAG.a.1367-b44","doi-asserted-by":"crossref","first-page":"665","DOI":"10.1038\/nmeth.1635","article-title":"Large-scale automated synthesis of human functional neuroimaging data","volume":"8","author":"Yarkoni","year":"2011","journal-title":"Nature Methods"},{"key":"2026091815555570700_IMAG.a.1367-b45","doi-asserted-by":"crossref","first-page":"1580808","DOI":"10.3389\/fnhum.2025.1580808","article-title":"Which software packages did researchers use to meta-analyze fMRI data? A literature survey from 2019 to 2024","volume":"19","author":"Yeung","year":"2025","journal-title":"Frontiers in Human Neuroscience"},{"key":"2026091815555570700_IMAG.a.1367-b46","article-title":"Deep sets","volume":"30","author":"Zaheer","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"key":"2026091815555570700_IMAG.a.1367-b47","doi-asserted-by":"crossref","first-page":"1783","DOI":"10.1109\/ISBI.2019.8759585","article-title":"fMRI data augmentation via synthesis","volume-title":"2019 IEEE 16th International Symposium on Biomedical Imaging (ISBI 2019)","author":"Zhuang","year":"2019"}],"container-title":["Imaging Neuroscience"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/direct.mit.edu\/imag\/article-pdf\/doi\/10.1162\/IMAG.a.1367\/2622849\/imag.a.1367.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/direct.mit.edu\/imag\/article-pdf\/doi\/10.1162\/IMAG.a.1367\/2622849\/imag.a.1367.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T19:56:09Z","timestamp":1789761369000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/imag\/article\/doi\/10.1162\/IMAG.a.1367\/138559\/Peaks2Image-Reconstructing-fMRI-maps-from"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":46,"URL":"https:\/\/doi.org\/10.1162\/imag.a.1367","relation":{},"ISSN":["2837-6056"],"issn-type":[{"value":"2837-6056","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026]]},"published":{"date-parts":[[2026]]},"article-number":"IMAG.a.1367"}}