Search

Purpose: We examine the rate of and reasons for follow-up in an Artificial Intelligence (AI)-based workflow for diabetic retinopathy (DR) screening relative to two human-based workflows.Patients and Methods: A DR screening program initiated September 2019 between one institution and its affiliated primary care and endocrinology clinics screened 2243 adult patients with type 1 or 2 diabetes without a diagnosis of DR in the previous year in the San Francisco Bay Area. For patients who screened positive for more-than-mild-DR (MTMDR), rates of follow-up were calculated under a store-and-forward human-based DR workflow ("Human Workflow"), an AI-based workflow involving IDx-DR ("AI Workflow"), and a two-step hybrid workflow ("AI–Human Hybrid Workflow"). The AI Workflow provided results within 48 hours, whereas the other workflows took up to 7 days. Patients were surveyed by phone about follow-up decisions.Results: Under the AI Workflow, 279 patients screened positive for MTMDR. Of these, 69.2% followed up with an ophthalmologist within 90 days. Altogether 70.5% (N=48) of patients who followed up chose their location based on primary care referral. Among the subset of patients that were seen in person at the university eye institute under the Human Workflow and AI–Human Hybrid Workflow, 12.0% (N=14/117) and 11.7% (N=12/103) of patients with a referrable screening result followed up compared to 35.5% of patients under the AI Workflow (N=99/279; χ2df=2 = 36.70, p < 0.00000001).Conclusion: Ophthalmology follow-up after a positive DR screening result is approximately three-fold higher under the AI Workflow than either the Human Workflow or AI–Human Hybrid Workflow. Improved follow-up behavior may be due to the decreased time to screening result. [ABSTRACT FROM AUTHOR]

Primary cilia are conserved organelles that integrate extracellular cues into intracellular signals and are critical for diverse processes, including cellular development and repair responses. Deficits in ciliary function cause multisystemic human diseases known as ciliopathies. In the eye, atrophy of the retinal pigment epithelium (RPE) is a common feature of many ciliopathies. However, the roles of RPE cilia in vivo remain poorly understood. In this study, we first found that mouse RPE cells only transiently form primary cilia. We then examined the RPE in the mouse model of Bardet-Biedl Syndrome 4 (BBS4), a ciliopathy associated with retinal degeneration in humans, and found that ciliation in BBS4 mutant RPE cells is disrupted early during development. Next, using a laser-induced injury model in vivo, we found that primary cilia in RPE reassemble in response to laser injury during RPE wound healing and then rapidly disassemble after the repair is completed. Finally, we demonstrated that RPE-specific depletion of primary cilia in a conditional mouse model of cilia loss promoted wound healing and enhanced cell proliferation. In summary, our data suggest that RPE cilia contribute to both retinal development and repair and provide insights into potential therapeutic targets for more common RPE degenerative diseases. [ABSTRACT FROM AUTHOR]

Purpose: To determine the prevalence of SARS-CoV-2 in tear samples and to investigate whether it correlates with ocular findings and patients' prognosis in Brazil. Methods: Tears were collected using Schirmer strips (SS) and conjunctival swabs (CS) from patients hospitalized with laboratory-confirmed SARS-CoV-2 infection. Samples were analyzed using qRT-PCR. Demographic and clinical data, ocular symptoms, and Schirmer tests (ST) were collected from patients. Charlson Comorbidity Index (CCI) was used to rate comorbidities, and patients were monitored until hospital discharge or death. Results: There were 61 hospitalized patients, 33 of which were diagnosed with COVID-19. Within the confirmed COVID-19 patients, SARS-CoV-2 was detected in 18.2% (n = 6) of CS and 12.1% (n = 4) of SS samples. Subjective and objective parameters for dry eye syndrome (e.g., ST COVID-19: 8.3 ± 6.4mm, non-COVID-19: 8.9 ± 6.6mm, p > 0.05) were comparable between COVID-19 (n = 33) and non-COVID-19 patients (n = 28). Among the 16 COVID-19 patients exhibiting ocular symptoms, only tearing was reported significantly more frequently when tear samples were positive for SARS-CoV-2 (p < 0.05). Strikingly, patients whose tears tested positive for SARS-CoV-2 had significantly inferior CCI (pos.: 34.0 ± 31.8%, neg.: 67.6 ± 36.4%, p < 0.05) and higher mortality rates (pos.: 50.0%, neg.: 7.4%, p < 0.01). Conclusions: SARS-CoV-2 was detected with a prevalence of 18.2% on the ocular surface. Decreased CCI and increased mortality rate in the positive tear group suggests that viral detection may relate to prognosis and highlight the need of personal protective measures for healthcare professionals. Most of the patients, regardless of COVID-19 diagnosis, had low tear production and eye discomfort, possibly pointing to the need for artificial tear use during hospitalization. [ABSTRACT FROM AUTHOR]

Introduction: Leber congenital amaurosis (LCA) type 2, due to disease‐causing variants in RPE65, is characterized by severe visual loss in early infancy. Current treatments include voretigene neparvovec‐rzyl (VN) for RPE65‐associated LCA. Herein, we present the long‐term follow‐up of a patient treated with VN using quantitative autofluorescence (488 nm excitation). Case Report: A 9‐year‐old girl with a diagnosis of LCA with biallelic variants in RPE65 presented for evaluation. The patient underwent VN treatment at the age of 11. The patient returned to clinic at age of 19 at which time imaging revealed evidence of chorioretinal atrophy. Quantitative autofluorescence performed prior to gene therapy and at 6‐ and 8‐year follow‐up revealed a central area of fundus autofluorescence. Discussion This case report demonstrates acquisition of fundus autofluorescence at 6‐ and 8‐year follow‐up despite the development of chorioretinal atrophy. [ABSTRACT FROM AUTHOR]

While color fundus photos are used in routine clinical practice to diagnose ophthalmic conditions, evidence suggests that ocular imaging contains valuable information regarding the systemic health features of patients. These features can be identified through computer vision techniques including deep learning (DL) artificial intelligence (AI) models. We aim to construct a DL model that can predict systemic features from fundus images and to determine the optimal method of model construction for this task. Data were collected from a cohort of patients undergoing diabetic retinopathy screening between March 2020 and March 2021. Two models were created for each of 12 systemic health features based on the DenseNet201 architecture: one utilizing transfer learning with images from ImageNet and another from 35,126 fundus images. Here, 1277 fundus images were used to train the AI models. Area under the receiver operating characteristics curve (AUROC) scores were used to compare the model performance. Models utilizing the ImageNet transfer learning data were superior to those using retinal images for transfer learning (mean AUROC 0.78 vs. 0.65, p-value < 0.001). Models using ImageNet pretraining were able to predict systemic features including ethnicity (AUROC 0.93), age > 70 (AUROC 0.90), gender (AUROC 0.85), ACE inhibitor (AUROC 0.82), and ARB medication use (AUROC 0.78). We conclude that fundus images contain valuable information about the systemic characteristics of a patient. To optimize DL model performance, we recommend that even domain specific models consider using transfer learning from more generalized image sets to improve accuracy. [ABSTRACT FROM AUTHOR]

Background: Inherited retinal dystrophies describe a heterogeneous group of retinal diseases that lead to the irreversible degeneration of rod and cone photoreceptors and eventual blindness. Recessive loss-of-function mutations in Tubulin Tyrosine Ligase Like 5 (TTLL5) represent a recently described cause of inherited cone-rod and cone dystrophy. This study describes the unusual phenotypes of three patients with autosomal recessive mutations in TTLL5. Examination of these patients included funduscopic evaluation, spectral-domain optical coherence tomography, short-wavelength autofluorescence, and full-field electroretinography (ffERG). Genetic diagnoses were confirmed using whole exome capture. Protein modeling of the identified variants was performed to explore potential genotype-phenotype correlations.Results: Genetic testing revealed five novel variants in TTLL5 in three unrelated patients with retinal dystrophy. Clinical imaging demonstrated features of sectoral cone-rod dystrophy and cone dystrophy, with phenotypic variability seen across all three patients. One patient also developed high-frequency hearing loss during a similar time period as the onset of retinal disease, potentially suggestive of a syndromic disorder. Retinal structure findings were corroborated with functional measures including ffERG findings that supported these diagnoses. Modeling of the five variants suggest that they cause different effects on protein function, providing a potential reason for genotype-phenotype correlation in these patients.Conclusions: The authors report retinal phenotypic findings in three unrelated patients with novel mutations causing autosomal recessive TTLL5-mediated retinal dystrophy. These findings broaden the understanding of the phenotypes associated with TTLL5-mediated retinal disease and suggest that mutations in TTLL5 should be considered as a potential cause of sectoral retinal dystrophy in addition to cone-rod and cone dystrophies. [ABSTRACT FROM AUTHOR]