Med-Banana：从成功与失败轨迹中学习质量可控的医学图像编辑
Med-Banana: Learning Quality-Controlled Medical Image Editing from Success-and-Failure Trajectories
AI总结：
提出Med-Banana框架，通过收集成功与失败编辑轨迹数据集Med-Banana-80K，联合训练编辑器、验证器和优化器，实现质量可控的医学图像编辑。
摘要：
文本引导的医学图像编辑必须满足所需的病理特征，同时保留解剖结构、模态特定外观和临床合理性。然而，现有数据集主要用最终接受的编辑结果来监督编辑器，并丢弃生成过程中产生的失败尝试。我们认为这些失败为质量控制提供了必要的监督：它们指定了应该拒绝什么、为什么编辑在医学或视觉上无效，以及应该如何修改指令。我们提出了Med-Banana，一个用于质量可控的医学图像编辑的轨迹监督框架。我们引入了Med-Banana-80K，一个大规模的成功与失败编辑轨迹资源，包含候选图像、验证结果、拒绝原因和提示优化。在此基础上，Med-Banana联合训练编辑器、验证器和优化器，实现了从接受和拒绝尝试中进行编辑-验证-优化推理。在MLLM评估者、盲审专家评估、源保留和真实-合成可分离性探测上的实验表明，与开放的医学图像编辑器相比，该方法具有一致的改进。代码和数据已公开。
英文摘要：
Text-guided medical image editing must satisfy the requested pathology while preserving anatomy, modality-specific appearance, and clinical plausibility. However, existing datasets largely supervise editors with final accepted edits and discard the failed attempts produced during generation. We argue that these failures provide essential supervision for quality control: they specify what should be rejected, why an edit is medically or visually invalid, and how the instruction should be revised. We present Med-Banana, a trajectory-supervised framework for quality-controlled medical image editing. We introduce Med-Banana-80K, a large-scale resource of success-and-failure editing trajectories with candidate images, verification outcomes, rejection reasons, and prompt refinements. Building on it, Med-Banana jointly trains an editor, verifier, and refiner, enabling edit--verify--refine inference from accepted and rejected attempts. Experiments across MLLM judges, blind expert assessment, source-preservation and real--synthetic separability probes demonstrate consistent improvements over open medical image editors. Code and data are publicly available.
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。
