arXiv cs.AI· Spiros Maggioros, Guray Erus, Gareth Harman, George Aidinis, Alexander Getka, Aristeidis Sotiras, Christos Davatzikos·· 12 小时前AI 评分31
Brain2Vec：面向结构性脑部 MRI 的自监督 3D 基础模型
Brain2Vec: A Self-Supervised 3D Foundation Model for Structural Brain MRI
AI 导读
Brain2Vec 是在 74,425 张 T1 加权 MRI 扫描上预训练的自监督 3D 基础模型，可作下游任务微调的权重初始化骨干，也可作独立特征提取器。用其预训练权重初始化并微调后，脑龄预测任务的最终精度提升、收敛更快；冻结嵌入在五项下游分类任务中的四项优于解剖学体积特征，并与在更大数据上预训练的 SOTA 基础模型表现相当。代码与模型已开源，也可通过 NiChart 云平台直接提取特征。
正文 · 原文
Abstract:Deep learning models have become essential tools in medical imaging, yet training them end-to-end for diverse tasks remains challenging due to limited labeled data and computational constraints. To address this, we introduce Brain2Vec, a foundation model pre-trained on 74,425 T1-weighted Magnetic Resonance Imaging (MRI) scans pooled from multiple studies. Brain2Vec can serve both as a weight initialization backbone for fine-tuning downstream tasks and as a standalone feature extractor. We evaluated Brain2Vec in three tasks: downstream predictive performance, convergence speed, and embedding quality. Using Brain2Vec's pretrained weights as initialization and fine-tuning on labeled data for a brain age prediction task, results demonstrate improved final accuracy and faster convergence compared to random initialization. Additionally, its frozen embeddings outperform anatomically defined volumetric features on four of five downstream classification tasks and perform competitively with a state-of-the-art foundation model pre-trained on substantially more data. Lastly, we show that Brain2Vec encodes information primarily from a spatially localized and neuroanatomically meaningful set of brain regions, and that manipulating a subset of latent features produces monotonic, region-specific effects on the reconstructed anatomy. Brain2Vec code and trained models are available as an open-source package on this https URL. Users can apply Brain2Vec locally by installing the package, or directly through the NiChart cloud platform this https URL, where Brain2Vec features can be derived without requiring local installation or specialized infrastructure.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.28411 [cs.CV] |
| (or arXiv:2610.28411v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.28411 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Spiros Maggioros [view email]
[v1]
Tue, 20 Oct 2026 15:50:35 UTC (6,313 KB)
来源：arXiv cs.AI · arxiv.org