专为MRI设计的分割模型，小结构分割更准，一键标注超快。
SAMRI: Segment Any MRI
- 只微调掩码解码器，节省94%训练时间，保留原始特征
- 对47种结构平均分割准确率达0.87，小结构提升42.4%
- 零样本泛化强，仅需4.5GB显存，适合临床快速部署
SAMRI是针对MRI优化的分割任意模型，通过框与点提示实现快速标注，显著提升全身体积MRI分割效果，尤其在小尺寸和临床关键结构上。现有SAM变体将MRI视为通用模态，忽视组织对比度差异、强度不均及小结构重要性。本研究提出专用基础模型，具备强大全身体积分割能力与零样本泛化性能。方法上，仅微调SAM（ViT-B/16）的掩码解码器，冻结编码器以保持预训练表征，减少94%训练时间、96%可训练参数及约99%计算量。训练使用110万张2D切片-掩码对，来自30个数据集，涵盖47种目标、T1/T2/FLAIR/DWI等多种对比度与全身体域。采用焦点Dice损失函数，结合边界框（可选点）提示。按掩码面积分层：小（<0.5%）、中（0.5%-3.5%）、大（>3.5%），显著性检验采用威尔科克斯符号秩检验。结果表明，使用框+点提示时，SAMRI平均DSC达0.87±0.11，优于MedSAM的0.74±0.24（提高17.6%，p<0.05），尤其在小结构（+42.4%）与中等结构（+26.9%）上优势明显。在六个零样本数据集上，平均DSC达0.85，优于基线模型。推理仅需约4.5GB显存，可通过标准硬件上的交互式界面运行。结论：基于大规模、专用于MRI的数据集进行解码器微调，可实现卓越的全身体积分割与强零样本泛化，尤其在小而关键结构上表现突出。公开代码、预训练模型与交互界面支持科研与临床应用。
原文摘要 · Abstract (English)
Summary: SAMRI is an MRI-specialized adaptation of the Segment Anything Model achieving superior whole-body MRI segmentation, particularly for small and clinically critical structures, through box and point prompts for rapid annotation. Purpose: Existing SAM adaptations treat MRI as a generic modality, overlooking variable tissue contrast, intensity inhomogeneity, and clinically important small structures. We propose an MRI-specialized foundation model with strong whole-body segmentation and zero-shot generalization for direct use on any MRI annotation task. Methods: SAMRI fine-tunes only the mask decoder of SAM (ViT-B/16), keeping encoders frozen to preserve pretrained representations and eliminate redundant passes-reducing training time by 94%, trainable parameters by 96%, and FLOPs by ~99% versus full-model retraining. Training used 1.1 million 2D slice-mask pairs from 30 datasets spanning 47 targets, T1/T2/FLAIR/DWI contrasts, and whole-body anatomy, with focal-Dice loss and bounding-box (with optional point) prompts. Sizes were stratified by mask area (small: <0.5%; medium: 0.5-3.5%; large: >3.5%), and significance assessed by the Wilcoxon signed-rank test. Results: SAMRI with box+point prompts achieved mean DSC 0.87 +/- 0.11 across 47 targets, outperforming MedSAM (0.74 +/- 0.24) by 17.6% (p < 0.05), with largest gains for small (+42.4%) and medium (+26.9%) structures. On six zero-shot datasets, SAMRI achieved mean DSC 0.85, outperforming baselines. Inference requires only ~4.5 GB VRAM through an interactive interface on standard hardware. Conclusion: Decoder-only fine-tuning on a large, MRI-specific corpus delivers superior whole-body segmentation with strong zero-shot generalization, particularly for small and clinically salient structures. Public code, pretrained models, and an interactive interface make SAMRI deployable for MRI segmentation research and clinical workflows.
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。