Deep learning and optical coherence tomography in glaucoma: Bridging the diagnostic gap on structural imaging

Area of Science:

• Ophthalmology
• Artificial Intelligence
• Medical Imaging

Background:

• Glaucoma is a primary cause of irreversible blindness globally.
• Optic nerve head damage in glaucoma can be visualized using optical coherence tomography (OCT).
• Early detection and monitoring are crucial for managing glaucoma and preventing vision loss.

Purpose of the Study:

• To review advancements in deep learning (DL) for glaucoma detection over the past decade.
• To explore the application of DL models trained on OCT data for improved diagnostic capabilities.
• To discuss the potential of DL in enhancing glaucoma detection using fundus photography.

Main Methods:

• Review of scientific literature on deep learning applications in glaucoma detection.
• Analysis of studies utilizing OCT imaging for automated glaucoma diagnosis and progression monitoring.
• Investigation of DL models trained on OCT data for fundus image analysis.

Main Results:

• Deep learning algorithms have shown significant improvements in automated glaucoma detection from OCT scans.
• DL models trained on OCT data enhance the accuracy of detecting glaucomatous damage on fundus photographs.
• These advancements expand the utility of accessible imaging modalities for glaucoma screening.

Conclusions:

• Deep learning has revolutionized glaucoma detection and monitoring, particularly with OCT data.
• DL models offer promising tools for early diagnosis and management of glaucoma.
• Future research should focus on translating these AI advancements into clinical practice and basic science research.