Background The efficiency of clinical trials for retinitis pigmentosa (RP) treatment is limited by the screening burden and lack of reliable surrogate markers for functional end points. Automated... Show moreBackground The efficiency of clinical trials for retinitis pigmentosa (RP) treatment is limited by the screening burden and lack of reliable surrogate markers for functional end points. Automated methods to determine visual acuity (VA) may help address these challenges. We aimed to determine if VA could be estimated using confocal scanning laser ophthalmoscopy (cSLO) imaging and deep learning (DL). Methods Snellen corrected VA and cSLO imaging were obtained retrospectively. The Johns Hopkins University (JHU) dataset was used for 10-fold cross-validations and internal testing. The Amsterdam University Medical Centers (AUMC) dataset was used for external independent testing. Both datasets had the same exclusion criteria: visually significant media opacities and images not centred on the central macula. The JHU dataset included patients with RP with and without molecular confirmation. The AUMC dataset only included molecularly confirmed patients with RP. Using transfer learning, three versions of the ResNet-152 neural network were trained: infrared (IR), optical coherence tomography (OCT) and combined image (CI). Results In internal testing (JHU dataset, 2569 images, 462 eyes, 231 patients), the area under the curve (AUC) for the binary classification task of distinguishing between Snellen VA 20/40 or better and worse than Snellen VA 20/40 was 0.83, 0.87 and 0.85 for IR, OCT and CI, respectively. In external testing (AUMC dataset, 349 images, 166 eyes, 83 patients), the AUC was 0.78, 0.87 and 0.85 for IR, OCT and CI, respectively. Conclusions Our algorithm showed robust performance in predicting visual impairment in patients with RP, thus providing proof-of-concept for predicting structure-function correlation based solely on cSLO imaging in patients with RP. Show less
Talib, M.; Cauwenbergh, C. van; Zaeytijd, J. de; Wynsberghe, D. van; Baere, E. de; Boon, C.J.F.; Leroy, B.P. 2021
Aim To investigate the natural history in a Belgian cohort of CRB1-associated retinal dystrophies. Methods An in-depth retrospective study focusing on visual function and retinal structure. Results... Show moreAim To investigate the natural history in a Belgian cohort of CRB1-associated retinal dystrophies. Methods An in-depth retrospective study focusing on visual function and retinal structure. Results Forty patients from 35 families were included (ages: 2.5-80.1 years). In patients with a follow-up of >1 year (63%), the mean follow-up time was 12.0 years (range: 2.3-29.2 years). Based on the patient history, symptoms and/or electroretinography, 22 patients (55%) were diagnosed with retinitis pigmentosa (RP), 15 (38%) with Leber congenital amaurosis (LCA) and 3 (8%) with macular dystrophy (MD), the latter being associated with the p.(Ile167_Gly169del) mutation (in compound heterozygosity). MD later developed into a rod-cone dystrophy in one patient. Blindness at initial presentation was seen in the first decade of life in LCA, and in the fifth decade of life in RP. Eventually, 28 patients (70%) reached visual acuity-based blindness (<0.05). Visual field-based blindness (<10 degrees) was documented in 17/25 patients (68%). Five patients (13%) developed Coats-like exudative vasculopathy. Intermediate/posterior uveitis was found in three patients (8%). Cystoid maculopathy was common in RP (9/21; 43%) and MD (3/3; 100%). Macular involvement, varying from retinal pigment epithelium alterations to complete outer retinal atrophy, was observed in all patients. Conclusion Bi-allelic CRB1 mutations result in a range of progressive retinal disorders, most of which are generalised, with characteristically early macular involvement. Visual function and retinal structure analysis indicates a window for potential intervention with gene therapy before the fourth decade of life in RP and the first decade in LCA. Show less