An artificial intelligence cloud platform for OCT-based retinal anomalies screening system in real clinical environments

Millions of individuals worldwide suffer from retinal anomalies, which can lead to irreversible vision loss. However, the number of ophthalmologists is highly mismatched with the population base in China, especially in many rural and underdeveloped towns. To tackle these challenges, this paper developed an Artificial Intelligence Cloud Platform for OCT-based Retinal Anomalies Screening (AI-PORAS), which is capable of detecting 15 retinal anomalies in OCT images to enhance remote diagnostic efficiency. AI-PORAS has been trained, validated, and deployed to 207 medical institutions in 29 provinces in China. The validation on 165,384 eyes with 3,551,959 OCT B-scan slices, AI-PORAS achieved an average accuracy of 93.16%, an AUC of 93.64%, a FPR of 6.82%, a FNR of 7.87%, matching the average performance level of attending ophthalmologists. Additionally, Statistical analysis of the 116,717 remotely diagnosed patients provided insightful guidance for healthcare decision-making and the development of tailored treatment plans.