Managing A Patient with Uveitis in The Era of Artificial Intelligence: Current Approaches, Emerging Trends, And Future Perspectives.

The integration of artificial intelligence (AI) with healthcare has opened new avenues for diagnosing, treating, and managing medical conditions with remarkable precision. Uveitis, a diverse group of rare eye conditions characterized by inflammation of the uveal tract, exemplifies the complexities in ophthalmology due to its varied causes, clinical presentations, and responses to treatments. Uveitis, if not managed promptly and effectively, can lead to significant visual impairment. However, its management requires specialized knowledge, which is often lacking, particularly in regions with limited access to care. AI's capabilities in pattern recognition, data analysis, and predictive modelling offer significant potential to revolutionize uveitis management. AI can classify disease outcomes, analyze multimodal imaging data, and identify new therapeutic targets. However, transforming these AI models into clinical applications and meeting patient expectations involves overcoming challenges like acquiring extensive, annotated datasets, ensuring algorithmic transparency, and validating these models in real-world settings. This review delves into the complexities of uveitis and the current AI landscape, discussing the development, opportunities, and challenges of AI from theoretical models to bedside application. It also examines the epidemiology of uveitis, the global shortage of uveitis specialists, and the disease's socioeconomic impacts, underlining the critical need for AI-driven approaches. Furthermore, it explores the integration of AI in diagnostic imaging and future directions in ophthalmology, aiming to highlight emerging trends that could transform management of a patient with uveitis and suggesting collaborative efforts to enhance AI applications in clinical practice.

Optical Coherence Tomography Features in Fovea-Off Exudative vs Rhegmatogenous Retinal Detachment.

PURPOSE: To describe the optical coherence tomography (OCT) features that can differentiate eyes with fovea-off exudative retinal detachment (ERD) vs. rhegmatogenous retinal detachment (RRD), with particular attention to outer retinal corrugations (ORCs). DESIGN: Multicenter, retrospective cross-sectional study. METHODS: Multicenter, retrospective cross-sectional study of patients diagnosed with unilateral or bilateral fovea-off ERD or primary, acute, fovea-off RRD between 2016 and 2021. This study was performed with the approval from the Research Ethics Board at the University of Toronto and was conducted in accordance with the Declaration of Helsinki. Patients with any ERD etiology and evidence of extensive, bullous fovea-off detachment and in the RRD group: consecutive patients with acute, primary fovea-off RRD with good quality baseline SD-OCT imaging were included. Patients with exudative choroidal neovascularization from any etiology, optic nerve pit, significant media opacity, or OCT images with poor quality or low signal strength were excluded. Primary outcome was to describe the morphological features of the macula using SD-OCT in patients diagnosed with ERD vs RRD, with specific interest in ORCs. RESULTS: 161 eyes (51 ERD and 110 RRD) of 154 patients were included. Fifty-one eyes with ERD presented with 1 of 15 etiologies. ERD were associated with a greater risk of having hyperreflective dots in the outer retina (92.2% vs 74.5%, p=0.009), hyperreflective material and dots in the subretinal fluid (72.5% vs 34.5%, p<0.001), internal limiting membrane and inner retinal undulations (70.6% vs 39.4%, p<0.001), and retinal pigment epithelium undulations (44.9% vs 6.4%, p<0.001) compared to RRD. RRD was associated with a greater risk of outer retinal corrugations (80% vs 0%, p<0.001), intraretinal fluid (90.9% vs 41.2%, p<0.001) and ellipsoid zone thickening (90% vs 66.7%, p<0.001) compared to ERD. CONCLUSION: The presence of ORCs are highly specific for RRD and absent in ERD. This is likely related to differences in the pathophysiology of the diseases process, specifically the content of the subretinal fluid. Understanding the differences in OCT morphological features of ERD vs RRD may aid with diagnosis and management.

Resolution of Postoperative Cystoid Macular Edema with a Suprachoroidal Triamcinolone Acetonide Injectable Suspension.

AIM: To report a case of significant postoperative cystoid macular edema (CME) in a patient with no prior history of uveitis or systemic inflammatory disease, in which a suprachoroidal (SC) triamcinolone acetonide injectable suspension resulted in complete resolution of CME. DESIGN: Case report. METHODS: An 81-year-old man presented with a complex, dense brunescent cataract, floppy iris, and miotic pupil in the left eye. Six weeks following cataract extraction, the intraocular lens (IOL) was dislocated completely from the visual axis. He underwent a pars plana vitrectomy and IOL exchange with an anterior chamber IOL due to a three-piece IOL dislocation. Following the operation, the patient began treatment with prednisolone acetate and ketorolac. However, 2 weeks later, he was noted to be a steroid responder and glaucoma suspect. Therefore, the decision was made to taper and discontinue topical steroids. At 6 months postoperatively, ranibizumab was injected for persistent CME. However, following injection, there was intraocular pressure (IOP) elevation and increased CME. IOP improved 1 month later with dorzolamide hydrochloride and timolol maleate. At 8 months post-operation, there was persistent CME. The patient was then treated with an SC triamcinolone acetonide injection. There was complete resolution of CME by 14 weeks later, with improvement in visual acuity. Despite the patient's known steroid response with topical steroids, IOP remained within normal limits during the entire follow-up period after injection. CONCLUSION: SC triamcinolone acetonide injection is indicated for macular edema associated with uveitis; however, treatment for postoperative macular edema with this novel suprachoroidal injection may be considered.

The Use of Large Language Models to Generate Education Materials about Uveitis.

OBJECTIVE: To assess large language models in generating readable uveitis information and in improving the readability of online health information. DESIGN: Evaluation of technology. SUBJECTS: Not applicable. METHODS: ChatGPT and Bard were asked the following prompts: (prompt A) "considering that the average American reads at a 6th grade level, using the Flesch-Kincaid Grade Level (FKGL) formula, can you write patient-targeted health information on uveitis of around 6th grade level?" and (prompt B) "can you write patient-targeted health information on uveitis that is easy to understand by an average American?" Additionally, ChatGPT and Bard were asked the following prompt from the first-page results of Google when the term "uveitis" was searched: "Considering that the average American reads at a 6th grade level, using the FKGL formula, can you rewrite the following text to 6th grade level: [insert text]." The readability of each response was analyzed and compared using several metrics described below. MAIN OUTCOME MEASURES: The FKGL is a highly validated readability assessment tool that assigns a grade level to a given text, the total number of words, sentences, syllables, and complex words. Complex words were defined as those with > 2 syllables. RESULTS: ChatGPT and Bard generated responses with lower FKGL scores (i.e., easier to understand) in response to prompt A compared with prompt B. This was only significant for ChatGPT (P < 0.0001). The mean FKGL of responses to ChatGPT (6.3 ± 1.2) was significantly lower (P < 0.0001) than Bard 10.5 ± 0.8. ChatGPT responses also contained less complex words than Bard (P < 0.0001). Online health information on uveitis had a mean grade level of 11.0 ± 1.4. ChatGPT lowered the FKGL to 8.0 ± 1.0 (P < 0.0001) when asked to rewrite the content. Bard was not able to do so (mean FKGL of 11.1 ± 1.6). CONCLUSIONS: ChatGPT can aid clinicians in producing easier-to-understand health information on uveitis for patients compared with already-existing content. It can also help with reducing the difficulty of the language used for uveitis health information targeted for patients. FINANCIAL DISCLOSURE(S): Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Anterior chamber inflammation grading methods: A critical review.

Assessing anterior chamber inflammation is highly subjective and challenging. Although various grading systems attempt to offer objectivity and standardization, the clinical assessment has high interobserver variability. Traditional techniques, such as laser flare meter and fluorophotometry, are not widely used since they are time-consuming. With the development of optical coherence tomography with high sensitivity, direct imaging offers an excellent alternative to assess objectively inflammation with the potential for automated analysis. We describe various anterior chamber inflammation grading methods and discuss their utility, advantages, and disadvantages.

Pathogen Surveillance for Acute Infectious Conjunctivitis.

Importance: Acute infectious conjunctivitis is a common ocular condition with major public health consequences. Objective: To assess regional variations and microbial etiologies of acute infectious conjunctivitis to guide treatment. Design, Setting, and Participants: In this cross-sectional study, patients with presumed acute infectious conjunctivitis were enrolled in the study at 5 sites (Honolulu, Hawaii; Los Angeles, San Francisco, and San Diego, California; and Petah-Tikva, Israel) from March 2021 to March 2023. Patients with allergic or toxic conjunctivitis were excluded. Main Outcomes and Measures: Pathogens were identified by unbiased RNA deep sequencing. Results: In all, 52 patients (mean [range] age, 48 [7-80] years; 31 females [60%]) were enrolled at 5 sites (6 patients from Honolulu, 9 from San Diego, 11 from Los Angeles, 13 from San Francisco, and 13 from Petah-Tikva). RNA deep sequencing detected human adenovirus species D in one-quarter of patients (13 of 52). A wide range of pathogens, including human coronavirus 229E, SARS-CoV-2, and herpes simplex virus type 1, was also identified, as well as several bacteria and fungi. Moreover, 62% (32 of 52) of patients presented with purulent discharge, while only 8% (4 of 52) of patients had confirmed bacterial pathogens. Conclusion and Relevance: In this cross-sectional study, pathogens associated with acute infectious conjunctivitis varied between all 5 sites in the US and Israel. Purulent discharge was a common presenting sign in this study, with a low specificity for bacteria-associated conjunctivitis, suggesting that further diagnostic workup may be necessary to inform antibiotic stewardship. Additional research on cost-effectiveness of using RNA deep sequencing is needed to ascertain whether it is better to monitor patients clinically until resolution of disease.

Comparison of Three Handheld Fundus Cameras for Assessment of the Vertical Cup-To-Disk Ratio.

Purpose To compare the quality of optic nerve photographs from three different handheld fundus cameras and to assess the reproducibility and agreement of vertical cup-to-disk ratio (VCDR) measurements from each camera. Methods Adult patients from a comprehensive ophthalmology clinic and an intravitreous injection clinic in northern Thailand were recruited for this cross-sectional study. Each participant had optic nerve photography performed with each of 3 handheld cameras: the Volk iNview, Volk Pictor Plus, and Peek Retina. Images were graded for VCDR in a masked fashion by two photo-graders and images with > 0.2 discrepancy in VCDR were assessed by a third photo-grader. Results A total of 355 eyes underwent imaging with three different handheld fundus cameras. Optic nerve images were judged ungradable in 130 (37%) eyes imaged with Peek Retina, compared to 36 (10%) and 55 (15%) eyes imaged with the iNview and Pictor Plus, respectively. For 193 eyes with gradable images from all 3 cameras, inter-rater reliability for VCDR measurements was poor or moderate for each of the cameras, with intraclass correlation coefficients ranging from 0.41 to 0.52. A VCDR ≥ 0.6 was found in 6 eyes on iNview images, 9 eyes on Pictor Plus images, and 3 eyes on Peek images, with poor agreement between cameras (e.g., no eyes graded as VCDR ≥ 0.6 on images from both the iNview and Pictor Plus). Conclusions Inter-rater reliability of VCDR grades from 3 handheld cameras was poor. Cameras did not agree on which eyes had large VCDRs.

Recurrent Anterior Uveitis Associated with Major Fluctuations in Choroidal Thickness in Patient with Pachychoroid Disorder.

PURPOSE: To describe remarkable choroidal thickness fluctuations corresponding to episodes of recurrent anterior uveitis with subretinal fluid development when exceeding a choroidal thickness threshold. METHODS: A patient with pachychoroid pigment epitheliopathy and unilateral acute anterior uveitis of the left eye was evaluated over a period of 3 years with multimodal retinal imaging including optical coherence tomography (OCT). Longitudinal changes in subfoveal choroidal thickness (CT) were measured and correlated with episodes of recurrent inflammation. RESULTS: Over the course of 5 recurrent episodes of inflammation in the left eye treated with oral antiviral and topical steroid therapy, subfoveal CT increased as much as 200 um or more. Subfoveal CT in the fellow quiescent right eye by contrast, was within normal limits and minimally changed throughout the follow up. Increased CT occurred with each episode of anterior uveitis and decreased by 200 µm or more during periods of quiescence in the affected left eye. Subretinal fluid and macular edema developed with a maximum CT of 468 um and spontaneously resolved when CT decreased after treatment. CONCLUSION: In eyes with pachychoroid disease, anterior segment inflammation may lead to marked increases in subfoveal CT and the development of subretinal fluid at a threshold thickness value.

Readability and Suitability of Online Uveitis Patient Education Materials.

PURPOSE: This study assesses readability, suitability, and accountability of online uveitis patient education materials (PEMs). METHODS: Two uveitis specialists reviewed the top 10 websites on Google for keyword "uveitis," with a PubMed review control. Readability was assessed using an online calculator, suitability using the Suitability Assessment of Materials (SAM) tool, and accountability using JAMA benchmarks. RESULTS: The mean SAM score was 21.05, indicating that websites were on average adequately suitable to educate patients. WebMD Uveitis website scored the highest at 25.5, while allaboutvision.org scored the lowest at 18.0. The average Flesch Reading Ease (FRE) score was 44.0 (95% CI: 34.2, 53.8). The average reading grade score was 11.0 (95% CI: 9.4, 12.6). WebMD Uveitis page scored highest on readability. The average accountability score among sites was 2.36 out of 4. CONCLUSIONS: Uveitis websites are only adequately suitable as PEMs, and all are above the recommended reading level. Uveitis specialists should advise patients on quality of online PEMs.

THE CHRYSANTHEMUM PHENOTYPE OF IDIOPATHIC MULTIFOCAL CHOROIDITIS.

PURPOSE: To describe the clinical characteristics and multimodal imaging features of a distinctive subtype of active idiopathic multifocal choroiditis (iMFC) lesions with grey-yellow chorioretinal lesions surrounded by smaller satellite dots, a presentation referred to as "chrysanthemum lesions." METHODS: Retrospective, observational, multicenter case series of eyes with active iMFC and chrysanthemum lesions. Multimodal imaging features were reviewed and presented. RESULTS: Twenty-five eyes from 20 patients (12 women and 8 men), with a mean age of 35.8 ± 17.0 years (range, 7-78 years) were included. Chrysanthemum lesions were equally located in the macula (48.0%) or the mid/far periphery (52.0%). The number of lesions per eye varied from 1 (16.0%) to more than 20 (56.0%). On optical coherence tomography, chrysanthemum lesions showed typical features of iMFC, including subretinal hyperreflective material splitting the retinal pigment epithelium/Bruch membrane. Chrysanthemum lesions were hypoautofluorescent on fundus autofluorescence imaging, hyperfluorescent on fluorescein angiography, hypofluorescent on indocyanine green angiography, and associated with choriocapillaris flow signal deficit on optical coherence tomography angiography. CONCLUSION: Active iMFC may present with findings resembling chrysanthemum lesions. The distinctive lesion morphology on ophthalmoscopic examination, the large number of lesions, and the high prevalence of exclusive midperipheral and far peripheral involvement may represent a distinctive phenotype of iMFC.

The role of a multicentre data repository in ocular inflammation: The Ocular Autoimmune Systemic Inflammatory Infectious Study (OASIS).

In the current literature, clinical registry cohorts related to ocular inflammation are few and far between, and there are none involving multi-continental international data. Many existing registries comprise administrative databases, data related to specific uveitic diseases, or are designed to address a particular clinical problem. The existing data, although useful and serving their intended purposes, are segmented and may not be sufficiently robust to design prognostication tools or draw epidemiological conclusions in the field of uveitis and ocular inflammation. To solve this, we have developed the Ocular Autoimmune Systemic Inflammatory Infectious Study (OASIS) Clinical Registry. OASIS collects prospective and retrospective data on patients with all types of ocular inflammatory conditions from centers all around the world. It is a primarily web-based platform with alternative offline modes of access. A comprehensive set of clinical data ranging from demographics, past medical history, clinical presentation, working diagnosis to visual outcomes are collected over a range of time points. Additionally, clinical images such as optical coherence tomography, fundus fluorescein angiography and indocyanine green angiography studies may be uploaded. Through the capturing of diverse, well-structured, and clinically meaningful data in a simplified and consistent fashion, OASIS will deliver a comprehensive and well organized data set ripe for data analysis. The applications of the registry are numerous, and include performing epidemiological analysis, monitoring drug side effects, and studying treatment safety efficacy. Furthermore, the data compiled in OASIS will be used to develop new classification and diagnostic systems, as well as treatment and prognostication guidelines for uveitis.

A Thematic Analysis of Online Uveitis Support Groups.

PURPOSE: To report the availability and activity of online uveitis support groups. METHODS: An online search was conducted for support groups for uveitis. Member count and activity were recorded. Posts and comments were graded along five themes: emotional or personal story sharing, information seeking, offer of outside information, emotional support, and expressions of gratitude. RESULTS: An online search resulted in 32 support groups for uveitis. Across all groups, there was a median membership of 725 (IQR 1410.5). Of the 32 groups, five were active and accessible at the time of study. In these five groups, 337 posts and 1406 comments were made within the past year. The most prevalent theme in posts consisted of information seeking (84%) while the most prevalent theme in comments consisted of emotion or personal story sharing (65%). CONCLUSIONS: Online uveitis support groups provide a unique space for emotional support, information sharing, and community building.Abbreviations: OIUF - Ocular Inflammation and Uveitis Foundation.