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

Defining Retinal Vasculitis

PURPOSE

To assess the validity of retinal vasculitis as the preferred diagnostic term for multiple conditions.

DESIGN

Perspective.

METHODS

This Perspective is based on expert opinion and review of literature focused on the current nosology and pathology of retinal vasculitis. Interpretation of the subset of intraocular inflammation named "retinal vasculitis" based on fundamental knowledge of the blood-retinal barrier, the neurovascular unit, and pathological and functional responses to a variety of stimuli is provided. Correlation with multimodal imaging and known mechanisms of immunologically mediated disease are discussed.

RESULTS

A search of Medline in early 2024 for the phrase "retinal vasculitis" resulted in 2041 citations encompassing immunologic, genetic, neoplastic, infectious, drug-related, and ischemia-related disorders. Classification schemes and angiographic grading systems are descriptive and do not address pathologic mechanisms adequately, in part due to lack of histologic confirmation. Although optical coherence tomography (OCT) angiography holds promise for better imaging of retinal vascular changes, it does not reveal the key feature of leakage and only partially improves understanding of pathophysiology. Diagnosing catastrophic retinal vascular occlusion after intravitreal injections as a retinal vasculitis is the most recent example of speculative application of the term to complex and rare disorders.

CONCLUSIONS

Retinal vasculitis is a diagnostic term that is over-used and imprecise. Revised nosology should limit the term to primary inflammation of the retinal vasculature itself that results in opening of the blood-retinal barrier with or without retinal vascular occlusions. Pending new histologic or mechanistic evidence, the provisional term of "retinal vascular inflammation" or "retinal vasculopathy" should be used for leakage or occlusion occurring in the context of intraocular inflammation.

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.