Fig. 6: Distribution and statistical agreements of choroidal metrics between SSOCT and Synthetically Enhanced SDOCT.
From: Optical coherence tomography choroidal enhancement using generative deep learning

The boxplots illustrate the details of the distribution of choroidal metrics calculated for SSOCT and synthetically enhanced SDOCT data (Synthetic) in the test set (146 eye pairs) for choroidal thickness, area, volume, and vascularity index; T tests were performed to find the significant difference between the groups in a pairwise comparison. All the boxplots represent the total range of values with interquartile ranges, where the central line indicates the median and the white dot defines the median. Error bars indicate the 95% confidence intervals. The scatter plots and Bland-Altman plots show the correlation of the choroidal metrics calculated between SSOCT and DL synthetically generated SDOCT data (Synthetic) in the test set (146 eye pairs). SDOCT Spectral-Domain Optical Coherence Tomography, SSOCT Swept-Source Optical Coherence Tomography, DR Diabetic Retinopathy, n.s. not significant, *** = P > 0.001.; a.u. = arbitrary unit.