可持续扩展的脑MRI基础模型

中文摘要 AI 辅助

脑磁共振成像（MRI）是神经科学和临床评估的核心，但现有模型通常针对特定疾病、人群或成像协议开发。基础模型有望提供更通用的表示，但它们通常是一次性预训练的，在使用新数据更新时可能会丢失早期能力。本文展示了Alcmaeon——一种三维脑MRI基础模型，在超过425,000个体积及衍生成像图谱上进行了无人工标签的预训练，可跨临床领域顺序扩展。Alcmaeon结合了体积编码、潜在扩散生成与Graph-Blueprint Pruning（GBP，图蓝图剪枝），该技术可保护对早期领域重要的网络模块，同时保留其余容量可训练。从健康衰老、神经退行性疾病向发育、精神疾病和肿瘤成像扩展时，在体素级重建指标上，GBP的遗忘程度低于顺序适配和弹性权重整合，且在适配肿瘤成像后优势最大。蓝图提供了可检查的记录，展示了模型容量如何被保护和复用。不同模型层级的表示支持图像合成、疾病分类、生存建模和术后预测，尽管没有单一表示对所有任务都是最优的。这些发现为脑MRI基础模型提供了一条途径，使其能随新出现的数据增长，同时保留早期能力。

英文摘要

Brain magnetic resonance imaging (MRI) is central to neuroscience and clinical assessment, but models are commonly developed for individual diseases, populations or imaging protocols. Foundation models promise more general representations, yet they are usually pretrained once and can lose earlier capabilities when updated with new data. Here we show that Alcmaeon, a three-dimensional brain MRI foundation model pretrained without manual labels on more than 425,000 volumes and derived imaging maps, can be expanded sequentially across clinical domains. Alcmaeon combines volumetric encoding and latent diffusion generation with Graph-Blueprint Pruning (GBP), which protects network modules important to earlier domains while leaving the remaining capacity trainable. Across expansion from healthy ageing and neurodegeneration to developmental, psychiatric and tumour imaging, GBP showed less forgetting than sequential adaptation and elastic weight consolidation across voxel-level reconstruction measures, with its largest advantage after adaptation to tumour imaging. The blueprints provided an inspectable record of how model capacity was protected and reused. Representations from different model levels supported image synthesis, disease classification, survival modelling and postoperative prediction, although no single representation was optimal for every task. These findings provide a route towards brain MRI foundation models that can grow with emerging data while retaining earlier capabilities.

发表机构

• University of Cambridge（剑桥大学）
• University of Málaga（马拉加大学）

机构由 AI 辅助整理，请以论文原文为准。