This study focuses on automating the diagnosis of glaucoma, a leading cause of irreversible vision loss, through machine learning classifiers (MLCs). Using data from 605 patients in Ghana, the research integrated Optical Coherence Tomography (OCT) and Visual Field Test (VFT) parameters to predict glaucoma status. Ten machine learning algorithms were tested, with Naïve Bayes achieving the highest diagnostic accuracy (AUROC 0.92). The findings suggest that machine learning can significantly improve diagnostic efficiency and accuracy, particularly in resource-constrained settings.

Traditional glaucoma diagnosis in Ghana faces challenges due to a high patient-to-clinician ratio, leading to long wait times and potential diagnostic delays. This study addresses the need for efficient, automated diagnostic tools to complement clinicians’ efforts, improving early detection and patient management.

The research primarily targets the healthcare sector, particularly ophthalmology and diagnostic technology industries. The findings are applicable to hospitals, eye clinics, and health systems aiming to enhance their diagnostic capabilities, especially in regions with high glaucoma prevalence.

The web application developed from this study offers potential for integration into existing hospital electronic medical records (EMR) systems. It can support faster patient processing, reduce diagnostic errors, and ensure consistent classification of glaucoma severity. This makes it a viable solution for healthcare facilities in Ghana and similar settings, improving operational efficiency and patient outcomes.

Adopt the Web Application: Integrate the developed diagnostic tool into hospital EMR systems for faster, more objective glaucoma diagnosis.
Further Development and Partnerships: Collaborate with technology developers and healthcare stakeholders to enhance features, including remote diagnosis and real-time data analysis.
Clinician Training: Develop training programs for clinicians on the effective use of AI-based diagnostic tools.
Validation and Scaling: Validate the tool with larger datasets and diverse populations to ensure reliability. Expand its use to other regions within Africa with similar glaucoma burdens.