Research Progress of Artificial Intelligence Image Analysis in Systemic Disease-Related Ophthalmopathy

Table 1

Summary of DR diagnosis model based on deep learning method.

Study	Task	Sample size	AI model	Output
Gulshan et al. [11]	Identification and detection	EyePACS-1 dataset and Messidor-2 dataset	Deep learning-trained algorithm	On the EyePACS-1 dataset, the AUC value was 0.991, the sensitivity was 90.3%, and the specificity was 98.1%; on the Messidor-2 dataset, the corresponding values were 0.990, 87.0%, and 98.5%.
Ai et al. [12]	Detection	35,126 images	DR-IIXRN	The AUC value and accuracy were 0.95 and 92%.
Bhardwaj et al. [13]	Identification and detection	The DRIVE, STARE, and DIARETDB1 datasets	InceptionResnet-V2	The accuracy was 93.33%.
Li et al. [14]	Detection	35201 images	Deep learning algorithm	The AUC value, sensitivity, and specificity were 0.989, 97.0%, and 91.4%, respectively
Li et al. [15]	Identification	120002 images	The retinal artificial intelligence diagnosis system	The accuracy was 98.1%.