arXiv 2304.039812023-08-29cs.LGcs.CV

不确定性启发的开放集学习用于视网膜异常识别

Uncertainty-inspired Open Set Learning for Retinal Anomaly Identification

Institute of High Performance Computing (IHPC), Agency for Science, Technology and Research (A*STAR)（新加坡科技研究局高性能计算研究所）；Medical College, Shantou University（汕头大学医学院）；College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics（南京航空航天大学计算机科学与技术学院）；College of Computer Science, Sichuan University（四川大学计算机学院）；School of Electronics and Information Engineering, Soochow University（苏州大学电子信息学院）；School of Biomedical Engineering, Anhui Medical University（安徽医科大学生物医学工程学院）；Longchuan People’s Hospital（龙川县人民医院）；Puning People’s Hospital（普宁市人民医院）；Haifeng PengPai Memory Hospital（海丰县彭湃纪念医院）；College of Intelligence and Computing, Tianjin University（天津大学智能与计算学部）；The Chinese University of Hong Kong（香港中文大学）；Soochow University（苏州大学）

Meng Wang, Tian Lin, Lianyu Wang, Aidi Lin, Ke Zou, Xinxing Xu, Yi Zhou, Yuanyuan Peng, Qingquan Meng, Yiming Qian, Guoyao Deng, Zhiqun Wu, Junhong Chen, Jianho… 展开作者

Meng Wang, Tian Lin, Lianyu Wang, Aidi Lin, Ke Zou, Xinxing Xu, Yi Zhou, Yuanyuan Peng, Qingquan Meng, Yiming Qian, Guoyao Deng, Zhiqun Wu, Junhong Chen, Jianhong Lin, Mingzhi Zhang, Weifang Zhu, Changqing Zhang, Daoqiang Zhang, Rick Siow Mong Goh, Yong Liu, Chi Pui Pang, Xinjian Chen, Haoyu Chen, Huazhu Fu

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

提出不确定性启发的开放集模型UIOS，通过不确定性分数识别未见类别样本，在多个数据集上显著优于标准AI模型，适用于视网膜异常筛查。

摘要：

在训练期间未能识别未见类别的样本，是人工智能在视网膜异常识别与分类实际应用中的主要局限性。我们建立了一个不确定性启发的开放集（UIOS）模型，该模型使用9种视网膜状况的眼底图像进行训练。除了评估每个类别的概率外，UIOS还计算了一个不确定性分数以表达其置信度。采用阈值策略的UIOS模型在内部测试集、外部目标类别（TC）-JSIEC数据集和TC未见测试集上分别取得了99.55%、97.01%和91.91%的F1分数，而标准AI模型的F1分数分别为92.20%、80.69%和64.74%。此外，UIOS正确预测了高不确定性分数，这将提示在非目标类别视网膜疾病、低质量眼底图像和非眼底图像数据集中需要进行人工检查。UIOS为视网膜异常的现实世界筛查提供了一种稳健的方法。

英文摘要：

Failure to recognize samples from the classes unseen during training is a major limitation of artificial intelligence in the real-world implementation for recognition and classification of retinal anomalies. We established an uncertainty-inspired open-set (UIOS) model, which was trained with fundus images of 9 retinal conditions. Besides assessing the probability of each category, UIOS also calculated an uncertainty score to express its confidence. Our UIOS model with thresholding strategy achieved an F1 score of 99.55%, 97.01% and 91.91% for the internal testing set, external target categories (TC)-JSIEC dataset and TC-unseen testing set, respectively, compared to the F1 score of 92.20%, 80.69% and 64.74% by the standard AI model. Furthermore, UIOS correctly predicted high uncertainty scores, which would prompt the need for a manual check in the datasets of non-target categories retinal diseases, low-quality fundus images, and non-fundus images. UIOS provides a robust method for real-world screening of retinal anomalies.

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。