Accuracy of Deep Learning Algorithms for the Diagnosis of Retinopathy of Prematurity by Fundus Images: A Systematic Review and Meta-Analysis

Figure 2

Performance of the DL models for detecting and grading ROP in primary analyses. Forest plots of sensitivities (a), specificities (b), and diagnostic odds ratios (DOR) (c), with respective confidence intervals, respectively, as well as to assess the heterogeneity in accuracy estimates across studies. Plots of individual study results in ROC space with receiver operating characteristics curve for all classifiers included (SROC) (d).

(a)

(b)

(c)

(d)