Search

BackgroundRetinal vascular diseases, including diabetic macular edema (DME), neovascular age-related macular degeneration (nAMD), myopic choroidal neovascularization (mCNV), and branch and central retinal vein occlusion (BRVO/CRVO), are considered vision-threatening eye diseases. However, accurate diagnosis depends on multimodal imaging and the expertise of retinal ophthalmologists. ObjectiveThe aim of this study was to develop a deep learning model to detect treatment-requiring retinal vascular diseases using multimodal imaging. MethodsThis retrospective study enrolled participants with multimodal ophthalmic imaging data from 3 hospitals in Taiwan from 2013 to 2019. Eye-related images were used, including those obtained through retinal fundus photography, optical coherence tomography (OCT), and fluorescein angiography with or without indocyanine green angiography (FA/ICGA). A deep learning model was constructed for detecting DME, nAMD, mCNV, BRVO, and CRVO and identifying treatment-requiring diseases. Model performance was evaluated and is presented as the area under the curve (AUC) for each receiver operating characteristic curve. ResultsA total of 2992 eyes of 2185 patients were studied, with 239, 1209, 1008, 211, 189, and 136 eyes in the control, DME, nAMD, mCNV, BRVO, and CRVO groups, respectively. Among them, 1898 eyes required treatment. The eyes were divided into training, validation, and testing groups in a 5:1:1 ratio. In total, 5117 retinal fundus photos, 9316 OCT images, and 20,922 FA/ICGA images were used. The AUCs for detecting mCNV, DME, nAMD, BRVO, and CRVO were 0.996, 0.995, 0.990, 0.959, and 0.988, respectively. The AUC for detecting treatment-requiring diseases was 0.969. From the heat maps, we observed that the model could identify retinal vascular diseases. ConclusionsOur study developed a deep learning model to detect retinal diseases using multimodal ophthalmic imaging. Furthermore, the model demonstrated good performance in detecting treatment-requiring retinal diseases.

BackgroundRetinal imaging has been applied for detecting eye diseases and cardiovascular risks using deep learning–based methods. Furthermore, retinal microvascular and structural changes were found in renal function impairments. However, a deep learning–based method using retinal images for detecting early renal function impairment has not yet been well studied. ObjectiveThis study aimed to develop and evaluate a deep learning model for detecting early renal function impairment using retinal fundus images. MethodsThis retrospective study enrolled patients who underwent renal function tests with color fundus images captured at any time between January 1, 2001, and August 31, 2019. A deep learning model was constructed to detect impaired renal function from the images. Early renal function impairment was defined as estimated glomerular filtration rate 6.5%, >7.5%, and >10%, respectively. ConclusionsThe deep learning model in this study enables the detection of early renal function impairment using retinal fundus images. The model was more accurate for patients with elevated serum HbA1c levels.

This review explored the role of mitochondria in retinal ganglion cells (RGCs), which are essential for visual processing. Mitochondrial dysfunction is a key factor in the pathogenesis of various vision-related disorders, including glaucoma, hereditary optic neuropathy, and age-related macular degeneration. This review highlighted the critical role of mitochondria in RGCs, which provide metabolic support, regulate cellular health, and respond to cellular stress while also producing reactive oxygen species (ROS) that can damage cellular components. Maintaining mitochondrial function is essential for meeting RGCs' high metabolic demands and ensuring redox homeostasis, which is crucial for their proper function and visual health. Oxidative stress, exacerbated by factors like elevated intraocular pressure and environmental factors, contributes to diseases such as glaucoma and age-related vision loss by triggering cellular damage pathways. Strategies targeting mitochondrial function or bolstering antioxidant defenses include mitochondrial-based therapies, gene therapies, and mitochondrial transplantation. These advances can offer potential strategies for addressing mitochondrial dysfunction in the retina, with implications that extend beyond ocular diseases. [ABSTRACT FROM AUTHOR]

Background: Premature children with retinopathy of prematurity (ROP) have been reported to an have increased risk of visual and neurocognitive impairments, yet little is known about whether vision could affect specific neurocognition. This study aimed to clarify the correlations between neurocognition and vision in premature children. Materials and Methods: This is a nonrandomized, cross-sectional, observational study in a pediatric cohort with five groups: (1) full-term (n = 25), (2) prematurity without ROP (n = 154), (3) prematurity with ROP but without treatment (n = 39), (4) prematurity with ROP and with bevacizumab (IVB) treatment (n = 62), and (5) prematurity with ROP and with laser/laser + IVB treatment (n = 20). Neurocognitive function was evaluated by the Wechsler Preschool and Primary Scale of Intelligence, Fourth Edition (WPPSI-IV) around the age of 4 years. Visual acuity (VA) and refractive errors were tested. Correlations between WPPSI parameters and visual outcomes were analyzed across five groups. Results: Among the 300 recruited children (mean age = 4.02 + 0.97 years, male = 56.3%), 297 were assessed by WPPSI-IV and 142 were assessed by vision tests. The Full-Scale Intelligence Quotient (FSIQ) index was worse in the premature groups. After adjusting for covariates, seven items, including FSIQ-Index (p = 0.047), fluid-reasoning index (p = 0.004), FR-percentile ranking (p = 0.008), object assembly (p = 0.034), picture concept (p = 0.034), zoo locations (p = 0.014) and bug search (p = 0.020), showed significant differences between groups. The better the best corrected VA (BCVA), the higher the scores on Verbal Comprehension Index (VCI), VCI-PR, and the subtest of information. Conclusions: Specific cognitive dysfunctions are related to the BCVA in this large cohort. Subtest performance profiles in WPPSI can be affected by prematurity, ROP treatment, and different ROP treatment. FSIQ is generally lower in premature children and even lower in children with ROP. [ABSTRACT FROM AUTHOR]

Aims: Sodium–glucose cotransporter‐2 inhibitors (SGLT2is) are proposed to alleviate the development of inflammatory eye diseases. However, the association between SGLT2i and retinal vascular occlusion remains unclear. Therefore, this study aims to explore the effects of SGLT2i on the incidence of retinal vascular occlusion. Materials and Methods: This retrospective cohort study analysed electronic medical records data from the largest multi‐institutional database in Taiwan. Individuals who initiated SGLT2is and dipeptidyl peptidase 4 inhibitors (DPP4is) between 2016 and 2019 were included in our analysis. To conduct a homogenous comparison, inverse probability of treatment weighting with propensity scoring was employed. The primary outcome was retinal vascular occlusion, and the secondary outcomes were retinal vascular occlusion‐related complications (macular oedema, vitreous haemorrhage, and tractional retinal detachment) and conditions requiring vitreoretinal intervention (intravitreal injection, retinal laser therapy, and vitrectomy). Results: In total, 12,074 SGLT2i users and 39,318 DPP4i users were included. The incidence rate of retinal vascular occlusion in the SGLT2i and DPP4i groups was 1.2 (95% confidence interval [CI], 0.9–1.4) and 1.6 (95% CI, 1.3–1.8) events per 1000 person‐years, respectively, which yielded a subdistribution hazard ratio (SHR) of 0.74 (95% CI, 0.55–0.99). Similar risk reductions were observed in the retinal vascular occlusion‐related complications (SHR, 0.76; 95% CI, 0.69–0.84) and conditions requiring vitreoretinal intervention (SHR, 0.84; 95% CI, 0.77–0.94). Conclusions: In this multi‐institutional study in Taiwan, SGLT2i use was associated with a reduced risk of retinal vascular occlusion. Further prospective studies are required to ascertain this association. [ABSTRACT FROM AUTHOR]

To assess the epidemiology, treatment, and outcomes of juvenile idiopathic arthritis (JIA)–associated uveitis (JIA-U) in Taiwan. Retrospective, multicenter database. Totally, 44 (6.1%) of the 722 JIA patients had uveitis. The mean ages of JIA and JIA-U diagnosis were 10.7 and 11.1 years, respectively. JIA-U patients had more antinuclear antibody presence. Among JIA-U patients, 25 (56.8%) were male, 11 (25.0%) experienced any ocular complication, and 8 (18.2%) received ocular surgery. Inactivity lasting ≥6 months was achieved in 17 (38.6%) patients; however, 11 (25.0%) of those experienced reactivation with a mean of 624.7 days after achieving inactivity. Female had more ocular complications and surgeries. Late age at JIA-U diagnosis and male predominance were distinctive in Taiwan. For patients with inactivity lasting ≥ 6 months was achieved, they still required close follow-up due to the high reactivation rate within 2 years. Female had poorer ocular outcomes. [ABSTRACT FROM AUTHOR]