曾 畅

磁共振成像（magnetic resonance imaging，MRI）具有多序列、多参数和高软组织分辨率等优势，但设备、场强和扫描协议差异也带来明显的数据异质性，传统人工智能模型因而常局限于特定序列、器官或任务。视觉基础模型、视觉—语言模型和医学大语言模型借助大规模预训练、三维表征及影像—文本联合学习，逐步用于图像重建、多序列融合、病灶识别、分子状态预测、报告生成和疗效评价。本文梳理MRI大模型的技术基础及其在精准诊疗中的应用，并讨论序列缺失、合成影像失真、跨中心泛化不足、无依据内容生成和临床证据有限等问题。现阶段，MRI大模型更适合承担图像质量控制、多序列信息整合和辅助决策，其临床价值仍需多中心前瞻性研究验证。

人工智能大模型；磁共振成像；多序列融合；影像表型；精准诊疗

[1] Sun Y,Wang L,Li G,et al.A foundation model for enhancing magnetic resonance images and downstream segmentation,registration and diagnostic tasks.Nat Biomed Eng,2025,9:521-538. [2] Tak D,Garomsa BA,Zapaishchykova A,et al.A generalizable foundation model for analysis of human brain MRI.Nat Neurosci,2026,29:945-956. [3] Cox J,Liu P,Stolte SE,et al.BrainSegFounder:Towards 3D foundation models for neuroimage segmentation.Med Image Anal,2024,97:103301. [4] Yang Z,DSouza N,Megyeri I,et al.Decipher-MR:a vision-language foundation model for 3D MRI representations.NPJ Digit Med,2026,9:419. [5] Park YW,Kang M,Ryu H,et al.A robust vision language model for molecular status prediction and radiology report generation in adult- type diffuse gliomas.NPJ Digit Med,2026,9:406. [6] Tejani AS,Klontzas ME,Gatti AA,et al.Checklist for Artificial Intelligence in Medical Imaging:2024 update.Radiol Artif Intell,2024,6:e240300. [7] Lekadir K,Feragen A,Fofanah AJ,et al.FUTURE-AI:international consensus guideline for trustworthy and deployable artificial intelligence in healthcare.BMJ,2025,388:e081554.

点击下载PDF