用30万张医学影像训练多模态多器官模型，提升分割精度。
Curia-MAE: Multi-Modal Multi-Anatomy MAE Pre-Training for 3D Medical Image Segmentation
- 基于卷积MAE改进重建、正则和相似性目标，增强预训练效果。
- 在8个分割任务上超越强基线，罕见病灶任务表现更优。
- 适合临床部署，冻结编码器可复用于多种疾病分割场景。
放射科基础模型通过仅微调顶层小模块即可迁移至新任务，但其评估仍以密集预测任务如3D分割为主，且当编码器冻结时，现有预训练模型仍落后于从头训练的nnU-Net。为缩小差距，本文扩展卷积MAE预训练，引入鲁棒重建目标、特征正则化与局部-全局相似性目标，提出Curia-MAE——一个在30万张涵盖多个解剖部位的CT和MRI图像上预训练的多模态、多器官模型。在8个聚焦解剖结构与病灶的分割基准测试中，该模型在冻结编码器设置下优于强基线，在全微调下保持竞争力，并在标签数据稀缺的病灶任务中表现更优。结果表明，单一冻结编码器可跨任务复用，显著降低临床应用中的模型适配与部署成本。预训练权重已公开发布于https://huggingface.co/raidium/Curia-MAE。
原文摘要 · Abstract (English)
Radiology foundation models learn transferable representations that can be adapted to new tasks by training only small layers on top of a frozen encoder. Dense prediction tasks such as 3D segmentation are, however, underrepresented in their evaluation, and, with the encoder kept frozen, pre-trained models still fall short of nnU-Net, the state-of-the-art reference trained from scratch. To close this gap we extend convolutional MAE pre-training with a robust reconstruction objective, a feature regularizer, and a local-global similarity objective. Using this method, we propose Curia-MAE, a multi-modal, multi-anatomy MAE model pre-trained on 300,000 CT and MRI images covering a large number of anatomical sites. On eight anatomy- and lesion-focused segmentation benchmarks, Curia-MAE improves frozen-encoder performance over a strong MAE baseline, while remaining competitive under full finetuning and superior on lesion tasks, where labeled data is scarce. These results indicate that a single frozen encoder can be reused across diverse segmentation tasks, reducing the cost of adapting and deploying such models in clinical workflows. Curia-MAE pre-trained model weights are made publicly available at https://huggingface.co/raidium/Curia-MAE.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。