医学影像分类与不确定性量化一体化模型，提升诊断可靠性。
MedSymmFlow: Bridging Generative Modeling and Classification in Medical Imaging through Symmetrical Flow Matching
- 基于对称流匹配构建生成-判别混合架构，统一分类与生成任务。
- 在四个MedMNIST数据集上分类准确率与AUC均优于或持平基线。
- 无需额外训练即可通过生成采样自然估算不确定性，适合临床决策支持。
可靠的医学图像分类需要准确的预测和校准良好的不确定性估计，尤其在高风险临床场景中至关重要。本文提出MedSymmFlow，一种基于对称流匹配的生成-判别混合模型，旨在统一医学影像中的分类、生成与不确定性量化。该模型采用潜在空间表示，可扩展至高分辨率输入，并引入语义掩码条件机制以增强诊断相关性。与标准判别模型不同，其通过生成采样过程天然实现不确定性估计。在涵盖多种模态和病灶的四个MedMNIST数据集上进行评估，结果表明MedSymmFlow在分类准确率与AUC方面达到或超过现有基线性能，同时在选择性预测下表现提升，验证了其不确定性估计的可靠性。
原文摘要 · Abstract (English)
Reliable medical image classification requires accurate predictions and well-calibrated uncertainty estimates, especially in high-stakes clinical settings. This work presents MedSymmFlow, a generative-discriminative hybrid model built on Symmetrical Flow Matching, designed to unify classification, generation, and uncertainty quantification in medical imaging. MedSymmFlow leverages a latent-space formulation that scales to high-resolution inputs and introduces a semantic mask conditioning mechanism to enhance diagnostic relevance. Unlike standard discriminative models, it naturally estimates uncertainty through its generative sampling process. The model is evaluated on four MedMNIST datasets, covering a range of modalities and pathologies. The results show that MedSymmFlow matches or exceeds the performance of established baselines in classification accuracy and AUC, while also delivering reliable uncertainty estimates validated by performance improvements under selective prediction.
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。