MedXAI: 一种结合检索增强和自我验证的框架,用于知识引导的医学图像分析
MedXAI: A Retrieval-Augmented and Self-Verifying Framework for Knowledge-Guided Medical Image Analysis
AI总结:
MedXAI通过整合深度学习与临床专家知识,提升医学图像分析的可解释性和泛化能力,尤其在稀有疾病检测中表现优异。
摘要:
准确且可解释的图像诊断仍然是医学人工智能中的基本挑战,特别是在领域转移和稀有类别条件下。深度学习模型往往难以应对现实中的分布变化,对罕见病理表现出偏见,并且缺乏在安全关键临床环境中所需的透明度。我们介绍了MedXAI(一种用于医学图像分类的可解释框架),这是一个基于专家知识的统一框架,它将深度视觉模型与临床专家的知识相结合,以提高泛化能力、减少稀有类别偏见,并通过局部化相关诊断特征来提供人类可理解的解释,而不是依赖技术后验方法(如显著性图、LIME)。我们评估了MedXAI在两种具有挑战性的任务上:(i)从静息状态fMRI中定位癫痫发作起始区,以及(ii)糖尿病视网膜病变分级。在十个多中心数据集上的实验显示了一致的提升,包括跨领域泛化能力提高了3%,稀有类别的F1分数提高了10%,显著优于强大的深度学习基线。消融分析确认了符号组件作为有效的临床先验和正则化器的作用,提高了在分布转移下的鲁棒性。MedXAI在提供临床对齐的解释的同时,实现了优于领域内和跨领域性能,特别是在多模态医学人工智能中的稀有疾病。
英文摘要:
Accurate and interpretable image-based diagnosis remains a fundamental challenge in medical AI, particularly under domain shifts and rare-class conditions. Deep learning models often struggle with real-world distribution changes, exhibit bias against infrequent pathologies, and lack the transparency required for deployment in safety-critical clinical environments. We introduce MedXAI (An Explainable Framework for Medical Imaging Classification), a unified expert knowledge based framework that integrates deep vision models with clinician-derived expert knowledge to improve generalization, reduce rare-class bias, and provide human-understandable explanations by localizing the relevant diagnostic features rather than relying on technical post-hoc methods (e.g., Saliency Maps, LIME). We evaluate MedXAI across heterogeneous modalities on two challenging tasks: (i) Seizure Onset Zone localization from resting-state fMRI, and (ii) Diabetic Retinopathy grading. Ex periments on ten multicenter datasets show consistent gains, including a 3% improvement in cross-domain generalization and a 10% improvmnet in F1 score of rare class, substantially outperforming strong deep learning baselines. Ablations confirm that the symbolic components act as effective clinical priors and regularizers, improving robustness under distribution shift. MedXAI delivers clinically aligned explanations while achieving superior in-domain and cross-domain performance, particularly for rare diseases in multimodal medical AI.
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。