Harnessing the power of artificial intelligence for glaucoma diagnosis and treatment

Abstract

Artificial intelligence (AI) has great potential for diagnosing and managing glaucoma, a disease that causes irreversible vision loss. Early detection is paramount to prevent visual field loss. AI algorithms demonstrate promising capabilities in analyzing various glaucoma investigations. In analyzing retinal fundus photographs, AI achieves high accuracy in detecting glaucomatous optic nerve cupping, a hallmark feature. AI can also analyze optical coherence tomography (OCT) images of the retinal nerve fiber layer(RNFL) and ganglion cell complex, identifying structural changes indicative of glaucoma and also Anterior Segment OCT(AS-OCT) for angle closure disease. OCT interpretation may even be extended to diagnose early features of systemic neurodegenerative diseases such as Alzheimer’s Disease and Parkinson’s Disease. Furthermore, AI can assist in interpreting visual field (VF) tests, including predicting future VF loss patterns for the next 5 years. The ability of AI to integrate data from multiple modalities, including fundus photographs, Intra Ocular Pressure(IOP) measurements, RNFL OCT, AS-OCT, and VF tests, paves the way for a more comprehensive glaucoma assessment. This approach has the potential to revolutionize ophthalmology by enabling teleophthalmology and facilitating the development of personalized treatment plans. However, the authors emphasize the crucial role of human judgement and oversight in interpreting AI-generated results. Ultimately, ophthalmologists must make the final decisions regarding diagnosis and treatment strategies.