biostudies-literatureWoof WANational Institute for Health Research (NIHR)Wellcome TrustNIHR Moorfields Biomedical Research Centre100652https://www.ebi.ac.uk/biostudies/studies/S-EPMC11782848biostudies-literatureUnknown5(2)<h4>Purpose</h4>To quantify relevant fundus autofluorescence (FAF) features cross-sectionally and longitudinally in a large cohort of patients with inherited retinal diseases (IRDs).<h4>Design</h4>Retrospective study of imaging data.<h4>Participants</h4>Patients with a clinical and molecularly confirmed diagnosis of IRD who have undergone 55° FAF imaging at Moorfields Eye Hospital (MEH) and the Royal Liverpool Hospital between 2004 and 2019.<h4>Methods</h4>Five FAF features of interest were defined: vessels, optic disc, perimacular ring of increased signal (ring), relative hypo-autofluorescence (hypo-AF), and hyper-autofluorescence (hyper-AF). Features were manually annotated by 6 graders in a subset of patients based on a defined grading protocol to produce segmentation masks to train an artificial intelligence model, AIRDetect, which was then applied to the entire imaging data set.<h4>Main outcome measures</h4>Quantitative FAF features, including area and vessel metrics, were analyzed cross-sectionally by gene and age, and longitudinally. AIRDetect feature segmentation and detection were validated with Dice score and precision/recall, respectively.<h4>Results</h4>A total of 45 749 FAF images from 3606 patients with IRD from MEH covering 170 genes were automatically segmented using AIRDetect. Model-grader Dice scores for the disc, hypo-AF, hyper-AF, ring, and vessels were, respectively, 0.86, 0.72, 0.69, 0.68, and 0.65. Across patients at presentation, the 5 genes with the largest hypo-AF areas were <i>CHM</i>, <i>ABCC6</i>, <i>RDH12</i>, <i>ABCA4</i>, and <i>RPE65</i>, with mean per-patient areas of 43.72, 29.57, 20.07, 19.65, and 16.92 mm², respectively. The 5 genes with the largest hyper-AF areas were <i>BEST1</i>, <i>CDH23</i>, <i>NR2E3</i>, <i>MYO7A</i>, and <i>RDH12</i>, with mean areas of 0.50, 047, 0.44, 0.38, and 0.33 mm², respectively. The 5 genes with the largest ring areas were <i>NR2E3, CDH23</i>, <i>CRX</i>, <i>EYS</i>, and <i>PDE6B</i>, with mean areas of 3.60, 2.90, 2.89, 2.56, and 2.20 mm², respectively. Vessel density was found to be highest in <i>EFEMP1</i>, <i>BEST1</i>, <i>TIMP3</i>, <i>RS1</i>, and <i>PRPH2</i> (11.0%, 10.4%, 10.1%, 10.1%, 9.2%) and was lower in retinitis pigmentosa (RP) and Leber congenital amaurosis genes. Longitudinal analysis of decreasing ring area in 4 RP genes (<i>RPGR</i>, <i>USH2A</i>, <i>RHO</i>, and <i>EYS</i>) found <i>EYS</i> to be the fastest progressor at -0.178 mm²/year.<h4>Conclusions</h4>We have conducted the first large-scale cross-sectional and longitudinal quantitative analysis of FAF features across a diverse range of IRDs using a novel AI approach.<h4>Financial disclosures</h4>Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.Ophthalmology scienceQuantification of Fundus Autofluorescence Features in a Molecularly Characterized Cohort of >3500 Patients with Inherited Retinal Disease from the United Kingdom.PMC11782848206619/Z/17/ZShah MFurman JMoosajee MSallum JLin SDownes SMWebster ARLiefers BJFu DJWoof WAMendes Bde Guimaraes TACBagga PPatel PLiu YBurke PGeorgiou MSousa da Silva AMahroo OAMadhusudhan SParry DHolz FGGhoshal BAl-Khuzaei SFujinami-Yokokawa YPontikos NMoghul IDaich Varela MNguyen QFujinami KMichaelides MSen SBalaskas KDe Silva SRSumodhee DLorenz BNaik GfalseQuantification of Fundus Autofluorescence Features in a Molecularly Characterized Cohort of >3500 Patients with Inherited Retinal Disease from the United Kingdom.<h4>Purpose</h4>To quantify relevant fundus autofluorescence (FAF) features cross-sectionally and longitudinally in a large cohort of patients with inherited retinal diseases (IRDs).<h4>Design</h4>Retrospective study of imaging data.<h4>Participants</h4>Patients with a clinical and molecularly confirmed diagnosis of IRD who have undergone 55° FAF imaging at Moorfields Eye Hospital (MEH) and the Royal Liverpool Hospital between 2004 and 2019.<h4>Methods</h4>Five FAF features of interest were defined: vessels, optic disc, perimacular ring of increased signal (ring), relative hypo-autofluorescence (hypo-AF), and hyper-autofluorescence (hyper-AF). Features were manually annotated by 6 graders in a subset of patients based on a defined grading protocol to produce segmentation masks to train an artificial intelligence model, AIRDetect, which was then applied to the entire imaging data set.<h4>Main outcome measures</h4>Quantitative FAF features, including area and vessel metrics, were analyzed cross-sectionally by gene and age, and longitudinally. AIRDetect feature segmentation and detection were validated with Dice score and precision/recall, respectively.<h4>Results</h4>A total of 45 749 FAF images from 3606 patients with IRD from MEH covering 170 genes were automatically segmented using AIRDetect. Model-grader Dice scores for the disc, hypo-AF, hyper-AF, ring, and vessels were, respectively, 0.86, 0.72, 0.69, 0.68, and 0.65. Across patients at presentation, the 5 genes with the largest hypo-AF areas were <i>CHM</i>, <i>ABCC6</i>, <i>RDH12</i>, <i>ABCA4</i>, and <i>RPE65</i>, with mean per-patient areas of 43.72, 29.57, 20.07, 19.65, and 16.92 mm², respectively. The 5 genes with the largest hyper-AF areas were <i>BEST1</i>, <i>CDH23</i>, <i>NR2E3</i>, <i>MYO7A</i>, and <i>RDH12</i>, with mean areas of 0.50, 047, 0.44, 0.38, and 0.33 mm², respectively. The 5 genes with the largest ring areas were <i>NR2E3, CDH23</i>, <i>CRX</i>, <i>EYS</i>, and <i>PDE6B</i>, with mean areas of 3.60, 2.90, 2.89, 2.56, and 2.20 mm², respectively. Vessel density was found to be highest in <i>EFEMP1</i>, <i>BEST1</i>, <i>TIMP3</i>, <i>RS1</i>, and <i>PRPH2</i> (11.0%, 10.4%, 10.1%, 10.1%, 9.2%) and was lower in retinitis pigmentosa (RP) and Leber congenital amaurosis genes. Longitudinal analysis of decreasing ring area in 4 RP genes (<i>RPGR</i>, <i>USH2A</i>, <i>RHO</i>, and <i>EYS</i>) found <i>EYS</i> to be the fastest progressor at -0.178 mm²/year.<h4>Conclusions</h4>We have conducted the first large-scale cross-sectional and longitudinal quantitative analysis of FAF features across a diverse range of IRDs using a novel AI approach.<h4>Financial disclosures</h4>Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.2025-01-01T00:00:00Z2025 Mar-Apr2026-05-26T20:12:04.985Z2025-04-05T23:33:58.866ZS-EPMC117828483989642210.1016/j.xops.2024.100652