A recent study in the EPMA Journal found that machine learning models based on structural, functional and vascular biomarkers—including those from optical coherence tomography (OCT) and OCT angiography (OCT-A)—can accurately predict the progression of primary open-angle glaucoma (POAG), paving the way for more personalized surveillance and preventive strategies in clinical practice.
“Glaucoma’s growing prevalence and persistent underdiagnosis produce a profound and multifaceted burden for patients, families, health care systems and societies worldwide, underscoring the critical importance of early detection, individualized monitoring of disease progression and effective interventions,” Natalia I. Kurysheva, of the Ophthalmological Center of the Federal and Medical Biological Agency of the Russian Federation, and colleagues wrote.
The researchers developed AI-based models to predict POAG progression using a broad panel of biomarkers drawn from OCT, OCT-A, automated perimetry and biomechanical assessments. OCT-A was included to detect unique insights into microvascular changes that can precede and help predict structural and functional decline. The goal of this study was to move beyond single metrics and capture the full spectrum of POAG heterogeneity that determines individual risk.
36-MONTH LONGITUDINAL STUDY OF 114 EYES
The researchers monitored 114 eyes across different stages of POAG for at least 36 months, then used the data to build predictive machine learning models using ranked partial least squares discriminant analysis.
They trained two prognostic models: one up to 27 parameters in early-stage POAG and 20 in advanced disease. Both models demonstrated high prognostic accuracy for classifying slow, moderate and rapid rates of POAG progression. The researchers found that different biomarkers dominate at different disease stages. In early stage POAG, retinal nerve fiber layer thickness, peripapillary microvascular dropout, parafoveal vascular density and corneal hysteresis emerged as the most important variables, the researchers found. For more advanced disease, age, ganglion cell complex thickness, specific macular thickness measures and peripapillary perfusion parameters were most predictive.
The study’s findings emphasize that vascular factors are significant predictors of glaucoma progression at both early and advanced disease stages. The researchers’ machine learning approach accommodates the complex, multifactorial progression of POAG and could enable clinicians to tailor follow-up intervals and focus on preventive interventions on patients at highest risk.
Author
