BrainNorm: A Foundation Model that knows Normal via Semantic Atlas Pretraining

• URL: http://arxiv.org/abs/2608.17521v1
• Date: Tue, 18 Aug 2026 08:45:02 GMT
• Title: BrainNorm: A Foundation Model that knows Normal via Semantic Atlas Pretraining
• Abstract summary: We introduce BrainNorm, a normative foundation model, trained and tested on 66,000 T1-weighted structural MRI (T1w sMRI) scans.<n>BrainNorm learns a Semantic Atlas Latent space (SAL), where each scan is represented as a set of atlas-parcel embeddings.<n>The resulting deviation patterns in SAL space enable zero-shot tasks for disease prediction using parcel-wise abnormalities.
• Score: 4.651750987298774
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We introduce BrainNorm, a normative foundation model, trained and tested on ~66,000 T1-weighted structural MRI (T1w sMRI) scans. By leveraging language-image style contrastive pretraining on healthy cohorts across ages, BrainNorm learns a Semantic Atlas Latent space (SAL), where each scan is represented as a set of atlas-parcel embeddings. This yields parcel-specific healthy aging template trajectories that support age-consistent template matching and localized deviation scoring relative to a subject's chronological age. Across 6 downstream cohorts, BrainNorm demonstrates generalization evaluated across 25 task-setting combinations spanning age estimation, brain-age gap estimation, parcel identification, and single- & multi-disease classification tasks under direct inference, zero-shot, few-shot & full-data linear-probe settings. The resulting deviation patterns in SAL space enable zero-shot tasks for disease prediction using parcel-wise abnormalities. Fine-tuning on healthy-only cohorts of downstream datasets further improves the performance of various tasks. Across all classification tasks, linear probing on BrainNorm's frozen embeddings outperforms 9 baselines finetuned under end-to-end supervision. Furthermore, the localized deviations identified by BrainNorm across various neurodegenerative disorders closely align with established neurodegeneration pathology in clinical literature.

Related papers

• GeoSAE: Geometric Prior-Guided Layer-Wise Sparse Autoencoder Annotation of Brain MRI Foundation Models [8.51943483922984]
  GeoSAE identifies a compact, fully interpretable feature set that predicts mild cognitive impairment (MCI)-to-AD conversion.<n>This shows that geometry-guided SAEs can extract interpretable, biomarkers from frozen brain MRI foundation models.
  arXiv  Detail & Related papers  (2026-05-03T11:42:25Z)
• Predicting Neuromodulation Outcome for Parkinson's Disease with Generative Virtual Brain Model [95.41752463487008]
  Parkinson's disease affects over ten million people worldwide.<n>Inter-individual variability limits empirical treatment selection, increasing non-negligible surgical risk and cost.<n>We bridge this gap with a pretraining-finetuning framework to predict outcomes directly from resting-state fMRI.
  arXiv  Detail & Related papers  (2026-03-31T02:36:10Z)
• Semi-disentangled spatiotemporal implicit neural representations of longitudinal neuroimaging data for trajectory classification [4.985839561141954]
  We present a new, fully data-driven method for representing aging trajectories across the entire brain by modelling subject-specific longitudinal T1-weighted MRI data as continuous functions using Implicit Neural Representations (INRs)<n>Our INR-based method achieves 81.3% accuracy for the brain aging trajectory classification task, outperforming a standard deep learning baseline model (73.7%)
  arXiv  Detail & Related papers  (2025-10-11T00:27:43Z)
• Conditional Fetal Brain Atlas Learning for Automatic Tissue Segmentation [0.8326370619658685]
  We introduce a novel deep-learning framework for generating continuous, age-specific fetal brain atlases.<n>The framework combines a direct registration model with a conditional discriminator.<n>It was trained on a curated dataset of 219 neurotypical fetal MRIs spanning from 21 to 37 weeks of gestation.
  arXiv  Detail & Related papers  (2025-08-06T15:07:39Z)
• Towards a general-purpose foundation model for fMRI analysis [58.06455456423138]
  We introduce NeuroSTORM, a framework that learns from 4D fMRI volumes and enables efficient knowledge transfer across diverse applications.<n>NeuroSTORM is pre-trained on 28.65 million fMRI frames (>9,000 hours) from over 50,000 subjects across multiple centers and ages 5 to 100.<n>It outperforms existing methods across five tasks: age/gender prediction, phenotype prediction, disease diagnosis, fMRI-to-image retrieval, and task-based fMRI.
  arXiv  Detail & Related papers  (2025-06-11T23:51:01Z)
• Naturalistic Language-related Movie-Watching fMRI Task for Detecting Neurocognitive Decline and Disorder [60.84344168388442]
  Language-related functional magnetic resonance imaging (fMRI) may be a promising approach for detecting cognitive decline and early NCD.<n>We examined the effectiveness of this task among 97 non-demented Chinese older adults from Hong Kong.<n>The study demonstrated the potential of the naturalistic language-related fMRI task for early detection of aging-related cognitive decline and NCD.
  arXiv  Detail & Related papers  (2025-06-10T16:58:47Z)
• Knowledge-Guided Prompt Learning for Lifespan Brain MR Image Segmentation [53.70131202548981]
  We present a two-step segmentation framework employing Knowledge-Guided Prompt Learning (KGPL) for brain MRI. Specifically, we first pre-train segmentation models on large-scale datasets with sub-optimal labels. The introduction of knowledge-wise prompts captures semantic relationships between anatomical variability and biological processes.
  arXiv  Detail & Related papers  (2024-07-31T04:32:43Z)
• Patched Diffusion Models for Unsupervised Anomaly Detection in Brain MRI [55.78588835407174]
  We propose a method that reformulates the generation task of diffusion models as a patch-based estimation of healthy brain anatomy. We evaluate our approach on data of tumors and multiple sclerosis lesions and demonstrate a relative improvement of 25.1% compared to existing baselines.
  arXiv  Detail & Related papers  (2023-03-07T09:40:22Z)
• Self-Supervised Longitudinal Neighbourhood Embedding [13.633165258766418]
  We propose a self-supervised strategy for representation learning named Longitudinal Neighborhood Embedding. Motivated by concepts in contrastive learning, LNE explicitly models the similarity between trajectory vectors across different subjects. We apply LNE to longitudinal T1w MRIs of two neuroimaging studies: a dataset composed of 274 healthy subjects, and Alzheimer's Disease Neuroimaging Initiative.
  arXiv  Detail & Related papers  (2021-03-05T17:55:53Z)
• ICAM-reg: Interpretable Classification and Regression with Feature Attribution for Mapping Neurological Phenotypes in Individual Scans [3.589107822343127]
  We take advantage of recent developments in generative deep learning to develop a method for simultaneous classification, or regression, and feature attribution. We validate our method on the tasks of Mini-Mental State Examination (MMSE) cognitive test score prediction for the Alzheimer's Disease Neuroimaging Initiative cohort. We show that the generated FA maps can be used to explain outlier predictions and demonstrate that the inclusion of a regression module improves the disentanglement of the latent space.
  arXiv  Detail & Related papers  (2021-03-03T17:55:14Z)
• 1-D Convlutional Neural Networks for the Analysis of Pupil Size Variations in Scotopic Conditions [79.71065005161566]
  1-D convolutional neural network models are trained for classification of short-range sequences. Model provides prediction with high average accuracy on a hold out test set.
  arXiv  Detail & Related papers  (2020-02-06T17:25:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.