Skip to main content
. Author manuscript; available in PMC: 2022 Feb 15.
Published in final edited form as: Optica. 2021 May 4;8(5):642–651. doi: 10.1364/optica.418274

Table 3.

Comparison among the Average Precision Scores of Different CNNs for Each Dataset and Statistical Analyses across All Subjectsa

Dataset WeakGCSeg UNet3D++ UNet3D VNet
IU 3.75° 0.90 ± 0.06 0.87 ± 0.06 0.89 ± 0.06 0.75 ± 0.06
IU 8.5° 0.91 ± 0.02 0.89 ± 0.03 0.90 ± 0.03 0.72 ± 0.02
IU 12.75° 0.90 ± 0.05 0.89 ± 0.05 0.86 ± 0.12 0.76 ± 0.06
FDA Healthy 0.92 ± 0.05 0.85 ± 0.05 0.85 ± 0.05 0.57 ± 0.19
FDA Glaucoma 0.77 ± 0.14 0.75 ± 0.14 0.74 ± 0.12 0.47 ± 0.18
Statistical analysis
Friedman’s avg. rank 1.29 (1st) 2.41 (3rd) 2.35 (2nd) 3.94 (4th)
Holm’s adjusted p — 0.0232 0.0232 6.8 × 10−9
Wilcoxon p — 0.0023 0.0086 2.9 × 10−4
a

Statistical tests are conducted at the subject level over all the data. Avg., average; p, p-value.