通过从 X-ray 专家模型蒸馏知识来引导胸部 CT 图像理解
Bootstrapping Chest CT Image Understanding by Distilling Knowledge from X-ray Expert Models
AI总结：
针对胸部 CT 缺乏大规模多病种标注数据的问题，本文提出从预训练 2D X-ray 专家模型蒸馏诊断知识，并结合语言引导检索与鲁棒对比学习，提升 3D CT 理解和报告生成能力。
摘要：
放射科医生非常渴望用于医学影像解读的全自动、多功能 AI。然而，缺乏经过广泛标注的大规模多病种数据集阻碍了这一目标的实现。本文探索了利用语言作为胸部 CT 成像天然高质量监督的可行性。鉴于图像-报告对的可用性有限，我们通过从经过广泛预训练的 2D X-ray 专家模型中蒸馏胸部相关诊断知识，来引导对 3D 胸部 CT 图像的理解。具体而言，我们提出一种语言引导的检索方法，将每张 3D CT 图像与其语义上最接近的 2D X-ray 图像匹配，并执行成对和语义关系知识蒸馏。随后，我们使用对比学习来对齐同一患者内的图像和报告，同时将它们与其他患者区分开。然而，当患者具有相似语义诊断（例如健康患者）时会出现挑战，若将其视为负样本可能造成混淆。我们引入了一种鲁棒对比学习，用于识别并纠正这些假负样本。我们使用超过 12,000 对胸部 CT 图像和放射学报告训练模型。在零样本学习、报告生成和微调过程等多种场景下的大量实验，证明了该模型在解读胸部 CT 图像方面的可行性。
英文摘要：
Radiologists highly desire fully automated versatile AI for medical imaging interpretation. However, the lack of extensively annotated large-scale multi-disease datasets has hindered the achievement of this goal. In this paper, we explore the feasibility of leveraging language as a naturally high-quality supervision for chest CT imaging. In light of the limited availability of image-report pairs, we bootstrap the understanding of 3D chest CT images by distilling chest-related diagnostic knowledge from an extensively pre-trained 2D X-ray expert model. Specifically, we propose a language-guided retrieval method to match each 3D CT image with its semantically closest 2D X-ray image, and perform pair-wise and semantic relation knowledge distillation. Subsequently, we use contrastive learning to align images and reports within the same patient while distinguishing them from the other patients. However, the challenge arises when patients have similar semantic diagnoses, such as healthy patients, potentially confusing if treated as negatives. We introduce a robust contrastive learning that identifies and corrects these false negatives. We train our model with over 12,000 pairs of chest CT images and radiology reports. Extensive experiments across multiple scenarios, including zero-shot learning, report generation, and fine-tuning processes, demonstrate the model's feasibility in interpreting chest CT images.
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。
