Using Deep Learning to Segment Retinal Vascular Leakage and Occlusion in Retinal Vasculitis.

PURPOSE: Retinal vasculitis (RV) is characterised by retinal vascular leakage, occlusion or both on fluorescein angiography (FA). There is no standard scheme available to segment RV features. We aimed to develop a deep learning model to segment both vascular leakage and occlusion in RV. METHODS: Four hundred and sixty-three FA images from 82 patients with retinal vasculitis were used to develop a deep learning model, in 60:20:20 ratio for training:validation:testing. Parameters, including deep learning architectures (DeeplabV3+, UNet++ and UNet), were altered to find the best binary segmentation model separately for retinal vascular leakage and occlusion, using a Dice score to determine the reliability of each model. RESULTS: Our best model for vascular leakage had a Dice score of 0.6279 (95% confidence interval (CI) 0.5584-0.6974). For occlusion, the best model achieved a Dice score of 0.6992 (95% CI 0.6109-0.7874). CONCLUSION: Our RV segmentation models could perform reliable segmentation for retinal vascular leakage and occlusion in FAs of RV patients.

RETINAL VASCULITIS SEVERITY ASSESSMENT: Intraobserver and Interobserver Reliability of a New Scheme for Grading Wide-Field Fluorescein Angiograms in Retinal Vasculitis.

PURPOSE: Wide-field fluorescein angiography is commonly used to assess retinal vasculitis (RV), which manifests as vascular leakage and occlusion. Currently, there is no standard grading scheme for RV severity. The authors propose a novel RV grading scheme and assess its reliability and reproducibility. METHODS: A grading scheme was developed to assess both leakage and occlusion in RV. Wide-field fluorescein angiography images from 50 patients with RV were graded by four graders, and one grader graded them twice. An intraclass correlation coefficient (ICC) was used to determine intraobserver-interobserver reliability. Generalized linear models were calculated to associate the scoring with visual acuity. RESULTS: Repeated grading by the same grader showed good intraobserver reliability for both leakage (ICC = 0.85, 95% CI 0.78-0.89) and occlusion (ICC = 0.82, 95% CI 0.75-0.88) scores. Interobserver reliability among four independent graders showed good agreement for both leakage (ICC = 0.66, 95% CI 0.49-0.77) and occlusion (ICC = 0.75, 95% CI 0.68-0.81) scores. An increasing leakage score was significantly associated with worse concurrent visual acuity (generalized linear models, ß = 0.090, P < 0.01) and at 1-year follow-up (generalized linear models, ß = 0.063, P < 0.01). CONCLUSION: The proposed grading scheme for RV has good to excellent intraobserver and interobserver reliability across a range of graders. The leakage score related to present and future visual acuity.

A systematic review of OCT and OCT angiography in retinal vasculitis.

BACKGROUND: Retinal vasculitis is a component of uveitis for which the Standardisation of Uveitis Nomenclature (SUN) working group has no standard diagnostic criteria or severity grading. Fluorescein angiography is the gold standard test to assess retinal vasculitis, but is invasive and time-consuming. Optical coherence tomography (OCT) provides non-invasive detailed imaging of retinal structures and abnormalities, including blood vessel architecture and flow with OCT angiography (OCT-A). However, use of OCT in retinal vasculitis beyond assessing macular oedema, is not well established. We conducted a systematic review to understand the features of retinal vasculitis in OCT, Enhanced-depth imaging OCT (OCT-EDI) and OCT-A imaging. METHODS: The systematic search was done in March 2022 and updated in January 2023, through PubMed, EMBASE and the Web of Science database for studies related to OCT, OCT-EDI and OCT-A findings and retinal vasculitis. Bias assessment was assessed using JBI Critical Appraisal Checklist, and any findings associated with retinal vasculitis were extracted by qualitative analysis. RESULTS: We identified 20 studies, including 8 articles on OCT, 6 on OCT-EDI and 6 on OCT-A. The studies included analytical retrospective studies, case-series, and a case-control study. Five OCT studies reported secondary complications could be detected, and four reported retinal thickness alteration in retinal vasculitis. Five studies explored choroidal thickness alteration in OCT-EDI, and four explored capillary density alterations in retinal vasculitis using OCT-A. The heterogeneity in the studies' analysis and design precluded a meta-analysis. DISCUSSION: There were no clear OCT, OCT-EDI or OCT-A findings that demonstrated potential to supersede fluorescein angiography for assessing retinal vasculitis. Some signs of macular structural effects secondary to retinal vasculitis may help prognostication for vision. The OCT signs of inflamed retinal vessels and perivascular tissue is an unexplored area.

Intercellular communication analysis of the human retinal pigment epithelial and choroidal cells predicts pathways associated with aging, cellular senescence and age-related macular degeneration.

The retinal pigment epithelium (RPE) and the choroid are ocular tissues with fundamental roles in supporting neuroretinal function. The pathogenesis of age-related macular degeneration (AMD), a leading cause of irreversible blindness for which aging is the highest risk factor is closely linked with progressive impairment of various functions of these tissues. Cellular senescence, marked by cell cycle arrest and secretion of proinflammatory factors, is known to be associated with aging and has been proposed as a potential driver of AMD. Here, we investigated the role played by intercellular communication in the RPE/choroid within the context of aging, senescence and AMD. We inferred cell-cell interactions in the RPE/choroid by applying CellChat and scDiffCom on a publicly available scRNA-seq dataset from three human donors with and without AMD. We identified age-regulated ligand and receptor genes by using limma on a separate publicly available bulk microarray dataset providing RPE/choroid samples at multiple time points. Cellular senescence was investigated by assigning a score to each cell and each sample of these scRNA-seq and microarray datasets, respectively, based on the expression of key signature genes determined by a previous senescence meta-analysis. We identified VEGF-, BMP-and tenascin-mediated pathways supporting some of the strongest cell-cell interactions between RPE cells, fibroblasts and choroidal endothelial cells and as strong intercellular communication pathways related to both aging and senescence. Their signaling strength was enhanced between subpopulations of cells having high senescence scores. Predominant ligands of these pathways were upregulated with age whereas predominant receptors were downregulated. Globally, we also observed that cells from AMD samples presented slightly bigger senescence scores than normal cells and that the senescence score positively correlated with age in bulk samples (R = 0.26, value of p < 0.01). Hence, our analysis provides novel information on RPE/choroid intercellular communication that gives insights into the connection between aging, senescence and AMD.

Visual outcome of endogenous endophthalmitis in Thailand.

To evaluate a 10-year visual outcome of endogenous endophthalmitis (EE) patients. A 10-year retrospective chart review of EE patients. Thirty-eight patients (40 eyes) were diagnosed with EE at the mean age of 42. Among the identifiable pathogens (71.1% culture positive), the causative agents were predominantly gram-negative bacteria (48.1%). The most common specie was Klebsiella pneumoniae (25.9%). About a quarter of the patients required surgical eye removal, and the remaining 45.7% had visual acuity (VA) worse than hand motion at one month after the infectious episode. The most common complication was ocular hypertension (52.5%). Poor initial VA was significantly associated with a worse visual outcome in the early post-treatment period (p 0.12, adjusted OR 10.20, 95% CI 1.65-62.96). Five patients continued to visit the clinic for at least ten years. One patient had gained his vision from hand motion to 6/7.5. Two patients had visual deterioration, one from corneal decompensation, and the other from chronic retinal re-detachment. Two patients developed phthisis bulbi, with either some VA perception of light or no light perception. Poor initial VA is the only prognostic factor of a poor early post-treatment visual outcome of EE.

Epidemiology of keratitis/scleritis-related endophthalmitis in a university hospital in Thailand.

To demonstrate the demographics, associated factors, clinical presentations, microbiology, management, visual outcome and complications of keratitis/scleritis-related endophthalmitis (KSE). A retrospective chart of all endophthalmitis patients diagnosed between September 2001 and August 2011 was reviewed. Only endophthalmitis cases with previous corneal or scleral infection were included in the study. The patients were followed until losing vision or eyeball, becoming phthisis, or the end of 2018. Eighty-seven patients with KSE were identified, all unilateral. The mean age was 56.4 ± 21.4 years. There was a slight male predilection (55 patients, 63.2%). The mean follow-up time was 50 ± 149 weeks. The causative pathogens were identified in 35 patients (40.2%), with the highest frequency being bacteria. The most common bacterium was Pseudomonas aeruginosa (n = 13), and the most common fungus was Aspergillus sp. (n = 5). Fifteen patients achieved (17.2%) final visual acuity (VA) of hand motion or better after treatment. Eyeball removal was performed in 61 (70.1%) patients. From multivariate analysis, the only prognostic factor for poor final VA (worse than hand motion, HM) was poor VA (worse than HM) at the initial visit (relative risk 1.97, 95% confidence interval 1.15-3.36, p = 0.013). KSE is uncommon but has a devastating outcome. We found that the patient's initial VA was the only predictor for their final vision. P. aeruginosa was the most common identifiable organism in this study. However, several fungal infections were recognised. These findings should raise awareness for treatment of KSE in the tropics.

Integrated Microarray and RNAseq Transcriptomic Analysis of Retinal Pigment Epithelium/Choroid in Age-Related Macular Degeneration.

We report for the first time an integrated transcriptomic analysis of RPE/choroid dysfunction in AMD (mixed stages) based on combining data from publicly available microarray (GSE29801) and RNAseq (GSE135092) datasets aimed at increasing the ability and power of detection of differentially expressed genes and AMD-associated pathways. The analysis approach employed an integrating quantitative method designed to eliminate bias among different transcriptomic studies. The analysis highlighted 764 meta-genes (366 downregulated and 398 upregulated) in macular AMD RPE/choroid and 445 meta-genes (244 downregulated and 201 upregulated) in non-macular AMD RPE/choroid. Of these, 731 genes were newly detected as differentially expressed (DE) genes in macular AMD RPE/choroid and 434 genes in non-macular AMD RPE/choroid compared with controls. Over-representation analysis of KEGG pathways associated with these DE genes mapped revealed two most significantly associated biological processes in macular RPE/choroid in AMD, namely the neuroactive ligand-receptor interaction pathway (represented by 30 DE genes) and the extracellular matrix-receptor interaction signaling pathway (represented by 12 DE genes). Furthermore, protein-protein interaction (PPI) network identified two central hub genes involved in the control of cell proliferation/differentiation processes, HDAC1 and CDK1. Overall, the analysis provided novel insights for broadening the exploration of AMD pathogenesis by extending the number of molecular determinants and functional pathways that underpin AMD-associated RPE/choroid dysfunction.

High incidence of invasive fungal infection during acute myeloid leukemia treatment in a resource-limited country: clinical risk factors and treatment outcomes.

BACKGROUND: Invasive fungal infection (IFI) causes high morbidity and mortality during acute myeloid leukemia (AML) treatment. Interventions to prevent fungal infection, including air filtration systems and antifungal prophylaxis, may improve outcomes in this group of patients. However, they are expensive and therefore inapplicable in resource-limited countries. The benefit of antifungal therapy is also dependent on the local epidemiology. That led us to conduct the study to evaluate the characteristics and impact of IFI in AML patients without prophylaxis in our setting. METHODS: Clinical data from patients with AML who have been treated with chemotherapy without antifungal prophylaxis were retrieved during a 5-year period at Thailand's hematology referral center. Incidence and risk factors of IFI and outcomes of patients were evaluated. RESULTS: Among 292 chemotherapy courses, there were 65 (22.3%) episodes of IFI. Of those, 10 (15.4%) were proven, 19 (29.2%) were probable, and 36 (55.4%) were categorized as being possible IFI. Molds were the most commonly observed causative pathogens (93.1%). The incidence of probable/proven IFI was highest during first induction (20.5%), followed by second induction (6.1%), and consolidation (2.7%). A long duration of neutropenia, old age, and low serum albumin were the strongest predictors of IFI. Compared with patients who had no IFI, patients with probable/proven IFI had a longer length of hospital stay and higher in-hospital mortality. Patients with proven IFI had a significantly worse outcome at 1 year. CONCLUSIONS: These results suggest the change in health policy to implement IFI preventive measures to improve outcomes of AML treatment.

Correction to: High incidence of invasive fungal infection during acute myeloid leukemia treatment in a resource-limited country: clinical risk factors and treatment outcomes.

Incorrect family name of Warissara Jutidamrongphan.