Figure 3 | Scientific Reports

Figure 3

From: Prediction of anti-vascular endothelial growth factor agent-specific treatment outcomes in neovascular age-related macular degeneration using a generative adversarial network

Figure 3

Comparison of the performance between the artificial intelligence model (AI), experienced human examiner (E1), and less experienced human examiner (E2) in predicting residual retinal fluid after initial treatment. In the ranibizumab group (A, n = 49), the sensitivity and specificity of the AI model are noted to be 0.615 and 0.667, respectively. In the aflibercept group (B, n = 49), the values are 0.857 and 0.881, respectively. Overall, the AI model shows comparable or relatively superior sensitivity and specificity than the human examiners.

Back to article page