Advancing Ocular Diagnostics: AI-based AS-OCT analysis to distinguish vitreoretinal lymphoma from uveitis

A comprehensive cross-sectional study explored the efficacy of Anterior Segment Optical Coherence Tomography (AS-OCT) in differentiating vitreous involvement between vitreoretinal lymphoma (VRL) and vitritis in uveitis (a form of eye inflammation).
Radiomic features were extracted from AS-OCT images of 28 patients and a classification model was constructed, achieving an impressive 87% ROC-AUC in correctly diagnosing VRL or uveitis.
The findings highlight the promising role of AS-OCT as a valuable adjunct to clinical assessments, providing ophthalmologists with a non-invasive tool for more precise and timely decision-making in the presence of suspected VRL.

Further developments are currently underway to implement a neural network capable of directly analyzing the tomographies, without the need for extracting radiomic features. This advancement aims to streamline the diagnostic process and enhance the accuracy of the assessments.