Quantification of Optical Coherence Tomography Features in >3500 Patients with Inherited Retinal Disease Reveals Novel Genotype-Phenotype Associations.
Woof W., de Guimarães TAC., Al-Khuzaei S., Varela MD., Shah M., Naik G., Sen S., Bagga P., Mendes B., Chan YW., Lin S., Ghoshal B., Liefers B., Fu DJ., Georgiou M., da Silva AS., Nguyen Q., Liu Y., Sumodhee D., Fujinami-Yokokawa Y., Patel PJ., Furman J., Moghul I., Moosajee M., Sallum J., De Silva SR., Lorenz B., Herrmann P., Holz FG., Fujinami K., Webster AR., Mahroo OA., Downes SM., Madhusuhan S., Balaskas K., Michaelides M., Pontikos N.
PURPOSE: To quantify spectral-domain optical coherence tomography (SD-OCT) images cross-sectionally and longitudinally in a large cohort of molecularly characterized patients with inherited retinal disease (IRDs) from the UK. DESIGN: Retrospective study of imaging data. PARTICIPANTS: Patients with a clinical and molecularly confirmed diagnosis of IRD who have undergone macular SD-OCT imaging at Moorfields Eye Hospital (MEH) between 2011 and 2019. We retrospectively identified 4,240 IRD patients from the MEH database (198 distinct IRD genes), including 69,664 SD-OCT macular volumes. METHODS: Eight features of interest were defined: retina, fovea, intraretinal cystic spaces (ICS), subretinal fluid (SRF), subretinal hyper-reflective material (SHRM), pigment epithelium detachment (PED), ellipsoid zone loss (EZ-loss) and retinal pigment epithelium loss (RPE-loss). Manual annotations of five b-scans per SD-OCT volume was performed for the retinal features by four graders based on a defined grading protocol. A total of 1,749 b-scans from 360 SD-OCT volumes across 275 patients were annotated for the eight retinal features for training and testing of a neural-network-based segmentation model, AIRDetect-OCT, which was then applied to the entire imaging dataset. MAIN OUTCOME MEASURES: Performance of AIRDetect-OCT, comparing to inter-grader agreement was evaluated using Dice score on a held-out dataset. Feature prevalence, volume and area were analysed cross-sectionally and longitudinally. RESULTS: The inter-grader Dice score for manual segmentation was ≥90% for retina, ICS, SRF, SHRM and PED, >77% for both EZ-loss and RPE-loss. Model-grader agreement was >80% for segmentation of retina, ICS, SRF, SHRM, and PED, and >68% for both EZ-loss and RPE-loss. Automatic segmentation was applied to 272,168 b-scans across 7,405 SD-OCT volumes from 3,534 patients encompassing 176 unique genes. Accounting for age, male patients exhibited significantly more EZ-loss (19.6mm 2 vs 17.9mm 2 , p<2.8×10 -4 ) and RPE-loss (7.79mm 2 vs 6.15mm 2 , p<3.2×10 -6 ) than females. RPE-loss was significantly higher in Asian patients than other ethnicities (9.37mm 2 vs 7.29mm 2 , p<0.03). ICS average total volume was largest in RS1 (0.47mm 3 ) and NR2E3 (0.25mm 3 ), SRF in BEST1 (0.21mm 3 ) and PED in EFEMP1 (0.34mm 3 ). BEST1 and PROM1 showed significantly different patterns of EZ-loss (p<10 -4 ) and RPE-loss (p<0.02) comparing the dominant to the recessive forms. Sectoral analysis revealed significantly increased EZ-loss in the inferior quadrant compared to superior quadrant for RHO (Δ=-0.414 mm 2 , p=0.036) and EYS (Δ=-0.908 mm 2 , p=1.5×10 -4 ). In ABCA4 retinopathy, more severe genotypes (group A) were associated with faster progression of EZ-loss (2.80±0.62 mm 2 /yr), whilst the p.(Gly1961Glu) variant (group D) was associated with slower progression (0.56 ±0.18 mm 2 /yr). There were also sex differences within groups with males in group A experiencing significantly faster rates of progression of RPE-loss (2.48 ±1.40 mm 2 /yr vs 0.87 ±0.62 mm 2 /yr, p=0.047), but lower rates in groups B, C, and D. CONCLUSIONS: AIRDetect-OCT, a novel deep learning algorithm, enables large-scale OCT feature quantification in IRD patients uncovering cross-sectional and longitudinal phenotype correlations with demographic and genotypic parameters.