自适应共形再分配用于医学图像分类中的类间过渡不确定性
Adaptive Conformal Redistribution for Inter-class Transitional Uncertainty in Medical Image Classification
- Indian Statistical Institute（印度统计学院）
- Jadavpur University（贾达普大学）
- University College London（伦敦大学学院）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
针对医学图像分类中过渡类别导致的模糊决策，提出自适应共形再分配（AdaConRed），通过五阶段流程将共形预测集转化为精细类别分配，在OSCC和ISIC基准上显著提升恶性类别准确率。
AI中文摘要：
医学图像分类常因过渡类别而变得复杂，这些类别的特征分布与相邻类别重叠，产生模糊的决策边界。共形预测返回具有不确定性意识的预测集，但这些在需要单一决策的临床筛查中不能直接应用。本工作提出自适应共形再分配（AdaConRed），一种无标签的后共形决策规则，将模糊的预测集转化为精细的类别分配。开发了一个五阶段流程。视觉-语言生成增强解决了少数类稀缺问题；冻结的DermFoundation编码器提供嵌入；轻量级多层感知器进行分类；熵调制、边缘感知的不一致性评分构建自适应预测集；预测为过渡性且具有多标签集的样本被重新分配到集合内最可能的替代类别，推理时仅使用模型输出。评估使用OSCC口腔病变和ISIC皮肤病变基准，误覆盖水平为0.2。在3类OSCC基准上，总体准确率从73.54%提高到77.38%，口腔癌准确率从64.29%提高到82.14%，良性准确率从56.57%提高到70.20%。过渡样本的重新分配使OPMD准确率从84.78%降至80.16%，与漏诊恶性肿瘤的不对称成本一致。在ISIC上，总体准确率从85.83%提高到87.19%，黑色素瘤准确率从66.04%提高到68.34%。在相同骨干和再分配规则下，AdaConRed优于LAC、APS和RAPS。共形预测可以超越不确定性量化，扩展到存在过渡性疾病类别的可操作决策支持，增益集中在临床关键的恶性类别。代码仓库：此https URL。
英文摘要：
Medical image classification is frequently complicated by transitional categories whose feature distributions overlap those of adjacent classes, producing ambiguous decision boundaries. Conformal prediction returns uncertainty-aware prediction sets, but these are not directly actionable in clinical screening, where a single decision is required. This work proposes adaptive conformal redistribution (AdaConRed), a label-free post-conformal decision rule that converts ambiguous prediction sets into refined class assignments. A five-stage pipeline is developed. Vision-language generative augmentation addresses minority-class scarcity; a frozen DermFoundation encoder provides embeddings; a lightweight multi-layer perceptron performs classification; an entropy-modulated, margin-aware nonconformity score constructs adaptive prediction sets; samples predicted as transitional with multi-label sets are reassigned to the most probable alternative class within the set, using only model outputs at inference. Evaluation uses the OSCC oral lesion and ISIC skin lesion benchmarks at a miscoverage level of 0.2. On the 3-class OSCC benchmark, overall accuracy improves from 73.54% to 77.38%, with oral cancer accuracy rising from 64.29% to 82.14% and benign accuracy from 56.57% to 70.20%. Reassignment of transitional samples reduces OPMD accuracy from 84.78% to 80.16%, consistent with the asymmetric cost of missed malignancy. On ISIC, overall accuracy improves from 85.83% to 87.19%, melanoma accuracy rising from 66.04% to 68.34%. AdaConRed outperforms LAC, APS and RAPS under an identical backbone and redistribution rule. Conformal prediction can be extended beyond uncertainty quantification toward actionable decision support where transitional disease categories are present, with gains concentrated in the clinically critical malignant categories. Code repository: https://github.com/saibal436ghosh/AdaConRed.
