CONFORM:
A Project to Create Crowd-Sourced Open Neuroscience fMRI Foundation Models

Foundation Models for the Brain and Body Workshop
Data on the Brain & Mind Tutorial Track
NeurIPS 2025
*Indicates Equal Contribution

#Indicates Equal Senior Contribution

CONFORM is built on these three amazing works:

CONFORM workflow. A single, optimized experimental design is distributed to multiple sites for data collection. The collected data is then centralized for preprocessing, alignment, and integration into a foundational dataset. This process creates a continuous feedback loop, allowing the dataset to grow in size and diversity, which informs future experimental design and provides the basis for a strong foundation model.

Abstract

We propose CONFORM (Crowd-Sourced Open Neuroscience fMRI Foundation Model), a project that will bring together recent advances in neural data processing and analysis with a novel, crowd-sourced infrastructure. This transformative approach will overcome several current challenges in creating a foundational human fMRI model for vision: collecting massive amounts of data from a handful of participants is neither scalable nor sustainable; the number of participants is small for such datasets; stimulus diversity is limited; and generalizability to different populations is poor. CONFORM will overcome these limitations by combining a powerful generative denoising method (SNAP), a scalable framework for aggregating existing fMRI datasets (MOSAIC), and a meta-learning model that enables generalization with much smaller data from new participants (BraInCoRL). Our collabo- rative effort will produce models built on unprecedented scale and diversity—ultimately with hundreds of participants and hundreds of thousands of naturalistic image and movie stimuli—and provide the tools for continuous expansion of the underlying dataset. This “crowd-sourced” approach will allow many more researchers to leverage state-of-the-art NeuroAI methods using the scale of data they typically collect, democratizing access to powerful models and accelerating scientific discovery for a wide range of neuroscientific domains and populations.

Workshop Paper

BibTeX


        @inproceedings{2025metalearning,
          title={Meta-Learning an In-Context Transformer Model of Human Higher Visual Cortex},
          author={Muquan Yu and Mu Nan and Hossein Adeli and Jacob S. Prince and John A. Pyles and Leila Wehbe and Margaret Marie Henderson and Michael J. Tarr and Andrew Luo},
          booktitle={The Thirty-ninth Annual Conference on Neural Information Processing Systems},
          year={2025},
          url={https://openreview.net/forum?id=B3iPTZh7Za}
        }