Interpretable Deep Learning for OCT-Based Diagnosis of Vitreoretinal Lymphoma Versus Uveitis. | ScholarDeck

Paper

Interpretable Deep Learning for OCT-Based Diagnosis of Vitreoretinal Lymphoma Versus Uveitis.

New research · Ophthalmology

Translational vision science & technology · 3mo

AI / informaticsTranslational vision science & technology· Feb 2026

Interpretable Deep Learning for OCT-Based Diagnosis of Vitreoretinal Lymphoma Versus Uveitis.

Azaz Khan, Ogul E Uner, Pengxiao Zang … Yali Jia

Read paper

Ocular OncologyAI / informaticsDiagnosisQ1IF 2.8

Deep learning model distinguishes eye lymphoma from uveitis using retinal scans.

Interpretable Deep Learning for OCT-Based Diagnosis of Vitreoretinal Lymphoma Versus Uveitis.

Azaz Khan … Yali Jia

Translational vision science & technology · Feb 2026 · Q1 in Ophthalmology

Purpose

To develop and validate an interpretable deep learning model that classifies optical coherence tomography (optical coherence tomography) scans as vitreoretinal lymphoma (vitreoretinal lymphoma) or non-infectious uveitis (non-infectious uveitis)-intermediate, posterior, or panuveitis-and visualizes differentiating pathological optical coherence tomography features.

Methods

This cross-sectional study included 45 patients with vitreoretinal lymphoma and 52 with non-infectious uveitis who underwent SPECTRALIS optical coherence tomography imaging.

n = 45 patients

76.0

Results

76.0

± 8.0

AUROC score - how well scans told lymphoma apart from uveitis

n = 45 patients

More results

Grad-CAM features of vitreoretinal lymphoma included preretinal deposits and changes in the retinal pigment epithelium.

“

Conclusion

A deep learning model with interpretability can reliably differentiate vitreoretinal lymphoma from non-infectious uveitis, highlighting disease-relevant optical coherence tomography features.

Read paper