丹娜-法伯癌症研究所Benjamin H. Kann团队取得一项新突破。他们的研究开发出了人脑MRI分析的可推广基础模型。2026年2月5日出版的《自然—神经科学》发表了这项成果。

在这里，研究小组提出了脑成像自适应核心（BrainIAC）——一个基础模型，旨在从未标记的脑MRI数据中学习广义表示，并作为各种下游应用适应的核心基础。在48965个广泛任务的大脑核磁共振成像上进行了训练和验证，研究小组证明了BrainIAC优于局部监督训练和其他预训练模型，特别是在低数据、少镜头、设置和高难度预测任务中，允许在其他不可行的场景中应用。BrainIAC可以集成到成像管道和多模式框架中，并可能导致改进的生物标志物发现和人工智能临床翻译。

研究人员表示，人工智能应用于脑磁共振成像（MRI）有可能推进神经系统疾病的诊断、预后和治疗计划。迄今为止，该领域一直受到有限的训练数据和特定任务模型的限制，这些模型不能很好地推广到患者群体和医疗任务中。通过利用自我监督学习、预训练和目标适应，基础模型为克服这些限制提供了一个有希望的范例。

附：英文原文

Title: A generalizable foundation model for analysis of human brain MRI

Author: Tak, Divyanshu, Garomsa, Biniam A., Zapaishchykova, Anna, Chaunzwa, Tafadzwa L., Climent Pardo, Juan Carlos, Ye, Zezhong, Zielke, John, Ravipati, Yashwanth, Pai, Suraj, Vajapeyam, Sri, Mahootiha, Maryam, Parker, Mitchell, Pike, Luke R. G., Smith, Ceilidh, Familiar, Ariana M., Liu, Kevin X., Prabhu, Sanjay, Arnaout, Omar, Bandopadhayay, Pratiti, Nabavizadeh, Ali, Mueller, Sabine, Aerts, Hugo JWL, Huang, Raymond Y., Poussaint, Tina Y., Kann, Benjamin H.

Issue&Volume: 2026-02-05

Abstract: Artificial intelligence applied to brain magnetic resonance imaging (MRI) holds potential to advance diagnosis, prognosis and treatment planning for neurological diseases. The field has been constrained, thus far, by limited training data and task-specific models that do not generalize well across patient populations and medical tasks. By leveraging self-supervised learning, pretraining and targeted adaptation, foundation models present a promising paradigm to overcome these limitations. Here we present Brain Imaging Adaptive Core (BrainIAC)—a foundation model designed to learn generalized representations from unlabeled brain MRI data and serve as a core basis for diverse downstream application adaptation. Trained and validated on 48,965 brain MRIs across a broad spectrum of tasks, we demonstrate that BrainIAC outperforms localized supervised training and other pretrained models, particularly in low-data, few-shot, settings and in high-difficulty prediction tasks, allowing for application in scenarios otherwise infeasible. BrainIAC can be integrated into imaging pipelines and multimodal frameworks and may lead to improved biomarker discovery and artificial intelligence clinical translation.

DOI: 10.1038/s41593-026-02202-6

Source: https://www.nature.com/articles/s41593-026-02202-6