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.
@article{arxiv.2304.03981,
title = {Uncertainty-inspired Open Set Learning for Retinal Anomaly Identification},
author = {Meng Wang and Tian Lin and Lianyu Wang and Aidi Lin and Ke Zou and Xinxing Xu and Yi Zhou and Yuanyuan Peng and Qingquan Meng and Yiming Qian and Guoyao Deng and Zhiqun Wu and Junhong Chen and Jianhong Lin and Mingzhi Zhang and Weifang Zhu and Changqing Zhang and Daoqiang Zhang and Rick Siow Mong Goh and Yong Liu and Chi Pui Pang and Xinjian Chen and Haoyu Chen and Huazhu Fu},
journal= {arXiv preprint arXiv:2304.03981},
year = {2023}
}