Anterior chamber inflammation grading with automatized image processing software

Anterior segment optical coherence tomography (AS-OCT) is a non-invasive imaging modality to assess anterior ocular structures at micron levels. AS-OCT image processing has been implemented in evaluating anterior chamber inflammation with differing approaches and biomarkers across studies. The primary objective of this study was to evaluate the agreement between AS-OCT–derived cell count and cell density and slit-lamp based anterior chamber cell grading using the Standardization of Uveitis Nomenclature (SUN) criteria. Secondarily, we aimed to integrate multiple AS-OCT–based biomarkers reported in prior studies into a single rule-based computational framework. We analyzed 450 images from 92 AS-OCT scans of 79 eyes (48 non-infectious, 10 infectious, 21 control) for cell count, density, morphology (area, eccentricty), aqueous-to-air relative intensity (ARI) index and anterior chamber (AC) area. Cell count and cell density were significantly different for each SUN grading score (p < 0.001). There was a significant positive correlation between the SUN grading score and ARI index (Spearman’s p < 0.001 rho = 0.410) and cell area (Spearman’s p < 0.001 rho = 0.22). Eyes with active inflammation demonstrated significantly higher ARI index (100.48[5.14] vs. 99.12[2.57] p < 0.001 r = 0.25, %95 CI [1.22, 2.71]) and eccentricity ( 0.57[0.19] vs. 0.00[0.75], p = 0.05, r = 0.11, %95 CI [0.00, 0.52]) than control eyes. No significant differences were found in cell morphology (area, eccentricity) between infectious and non-infectious uveitis. Automated AS-OCT based image processing provided a rule-based, device-specific computational framework for quantifying AC inflammation that alings with clinical grading and it may serve as a complementary tool for assessing anterior chamber inflammation in uveitis.