TRIAGE：面向标注高效的半监督视网膜OCT分类的风险控制伪标签准入框架
TRIAGE: Risk-Controlled Pseudo-Label Admission for Annotation-Efficient Semi-Supervised Retinal OCT Classification
浏览论文内容
中文总结 AI 辅助
TRIAGE是采用患者级保形风险控制器等模块的半监督框架，仅用少量标注数据就实现了视网膜OCT分类的优异性能，且显著优于现有同类方法。
中文摘要 AI 辅助
先进的视网膜疾病诊断成像模态光学相干断层扫描（OCT）面临自动化不足的问题，原因在于专家标注的成本高昂。半监督学习（SSL）可利用未标注的B-scan解决标注不足的问题，但当前大多数生成伪标签的技术仅基于预测置信度，未考虑不同类型错误之间的不对称性。本文提出TRIAGE，这是一种面向OCT扫描分类的风险控制半监督框架，采用患者水平保形风险控制器与不对称代价矩阵的概念。TRIAGE整合了三个关键模块：能够处理疾病亚型部分异常监督的分层分类器、带有原始对偶覆盖控制的患者分组保形风险控制器，以及用于跨切片验证的上下文感知Transformer教师模型。在来自Noor眼科医院的数据集（含16822张B-scan、161名患者、554个体积）中，针对未见过的患者测试集，仅使用20%标注数据时，TRIAGE实现了89.66%的扫描水平准确率、0.8805的宏F1值、0.9641的宏AUC值，以及8.34%的欠分级率；仅使用5%标注数据时，TRIAGE仍保持76.88%的准确率和0.1656的欠分级率。与其他六种最先进的半监督方法相比，TRIAGE表现显著更优，消融研究证明了每个模块对整体框架性能的贡献（与固定阈值方法相比，欠分级率降低42.7%）。在OCT-C8数据集上，使用1%标注数据时，TRIAGE在3分类任务中达到98.00%的准确率；使用10%标注数据时，在8分类任务中达到95.94%的准确率。
英文摘要
The advanced retinal disease diagnosing imaging modality, optical coherence tomography (OCT), encounters a lack of automation because of the high expenses for annotations performed by specialists. The use of SSL solves the problem of insufficient annotations using unlabeled B-scans; however, most of the current techniques for generating pseudo-labels are based on prediction confidence without considering the asymmetry between different types of errors. This paper proposes TRIAGE, a risk-controlled semi-supervised framework for OCT scans classification, which uses the concept of a patient-level conformal risk controller with an asymmetric cost matrix. TRIAGE unites three crucial modules: a hierarchical classifier that is capable of working with partially abnormal supervision of the disease subtypes, a patient-grouped conformal risk controller with primal-dual coverage control, and a context-aware Transformer teacher for cross-slice verification. On the dataset from Noor Eye Hospital (16,822 B-scans, 161 patients, and 554 volumes) with a test set of unseen patients, TRIAGE demonstrates 89.66% scan-level accuracy, 0.8805 macro-F1, 0.9641 macro-AUC, and an 8.34% under-grading rate when using only 20% of the labeled data. With only 5% of the labeled data, TRIAGE keeps 76.88% accuracy and a 0.1656 under-grading rate. Compared with the other six state-of-the-art semi-supervised methods, TRIAGE significantly outperforms them with ablation study demonstrating the contribution of each module in the overall framework performance (by 42.7% in terms of under-grading rate comparing to fixed threshold methods). TRIAGE demonstrates 98.00% accuracy for 3-class classification with 1% labeled data and 95.94% accuracy for 8-class classification with 10% labeled data on the OCT-C8 dataset.
发表机构
- Rajshahi University of Engineering & Technology（拉杰沙希工程技术大学）
- Elite Research Lab LLC（精英研究实验室有限责任公司）
- Multimedia University（多媒体大学）
机构由 AI 辅助整理，请以论文原文为准。
