扩散细化分割与视觉-语言解释用于儿童脑肿瘤MRI
Diffusion-Refined Segmentation and Vision-Language Interpretation for Pediatric Brain Tumor MRI
AI总结：
提出两阶段框架，先用Swin-UNETR粗分割，再用条件扩散模型细化边界，最后结合多模态语言模型生成结构化报告，提升儿童脑肿瘤分割精度和可解释性。
摘要：
由于标注数据有限、成像表型异质性、肿瘤边界弥散以及肿瘤子区域类别不平衡，准确的儿童脑肿瘤分割仍然具有挑战性。在此，我们提出一个两阶段深度学习框架，用于改进多模态儿童脑MRI分割和临床解释。首先，我们在BraTS-PEDs MRI扫描上评估3D Res U-Net和Swin-UNETR基线模型，使用四种配准模态预测肿瘤核心、全肿瘤和增强肿瘤区域。其次，我们引入基于扩散的细化模型，以粗Swin-UNETR预测为条件，包括3D DDPM细化器和MedSegDiff。条件化显著提高了扩散稳定性和性能，特别是对于增强肿瘤边界分割。条件化MedSegDiff实现了最强的边界一致性，HD95最低。最后，预测的肿瘤体积和代表性分割叠加图与多模态语言模型集成，生成结构化的放射学风格报告。综合来看，我们的结果表明，从粗到细的扩散分割可以改善儿童肿瘤边界描绘，并支持端到端可解释的AI辅助神经肿瘤学工作流程。
英文摘要：
Accurate pediatric brain tumor segmentation remains challenging due to limited annotated data, heterogeneous imaging phenotypes, diffuse tumor boundaries, and class imbalance across tumor subregions. Here, we present a two-stage deep learning framework for improving multi-modal pediatric brain MRI segmentation and clinical interpretation. First, we evaluate 3D Res U-Net and Swin-UNETR baselines on BraTS-PEDs MRI scans, using four co-registered modalities to predict tumor core, whole tumor, and enhancing tumor regions. Second, we introduce diffusion-based refinement models conditioned on coarse Swin-UNETR predictions, including a 3D DDPM refiner and MedSegDiff. Conditioning substantially improves diffusion stability and performance, particularly for enhancing tumor boundary segmentation. Conditioned MedSegDiff achieves the strongest boundary agreement with the lowest HD95. Finally, predicted tumor volumes and representative segmentation overlays are integrated with a multimodal language model to generate structured radiology-style reports. Together, our results suggest that coarse-to-refined diffusion segmentation can improve pediatric tumor boundary delineation and support end-to-end interpretable AI-assisted neuro-oncology workflows.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
