Digital slit lamp derived parameters in shallow anterior chamber screening

To evaluate anterior chamber depth (ACD) using diffraction parameters derived from slit-lamp anterior segment (AS) images and assess their diagnostic efficacy in identifying optimal cutoff values for shallow ACD screening. In this cross-sectional study, from September 21, 2022 to June 30, 2023, 75 volunteers (150 eyes). Slit-lamp images (450 images, 3 pictures per eye) were taken, and the ACD was determined using an IOL Master 700. After manual annotation of the anatomical landmarks, the anterior chamber cross-sectional area (ACCSA) and anterior chamber angle (ACA) of the slit-lamp images were measured. Pixel was applied after image processing and feature extraction to predict the ACD. These values were then compared with the ACD acquired from the IOLMaster 700. The correlation between the parameters of the two techniques was determined by Pearson correlation analysis. The anterior chamber area (ACA) under the receiver operating characteristic curve (AUC) was calculated to evaluate the diagnostic efficacy of relevant indicators on images of the ACD. A total of 450 photographs(207 shallow AC and 243 deep AC images) were analyzed. The nasal/temporal anterior chamber cross-sectional area (N/TACCSA), the sum of the cross-sectional areas of the central and nasal/temporal anterior chambers (SACCSA), and nasal/temporal anterior chamber angle (N/TACA) achieved better performance (AUC = 0.819, 95% confidence interval (CI) = 0.752-0.887; AUC = 0.818, 95% CI = 0.748-0.887 and AUC = 0.833, 95% CI = 0.766-0.899, respectively). The image parameter values were strongly positively correlated with the ACD of IOLMaster 700 (central anterior chamber cross-sectional area (CACCSA, N/TACCSA, SACSA, and N/TACA, R = 0.603, 0.651, 0.645, and 0.665, P < 0.01, respectively). Some image parameters not only had a strong positive correlation with the measured ACD values from the IOLMaster 700 but also had the ability to screen for shallow ACD. This study suggested the possibility of estimating the central ACD using AS images.