Age-related macular degeneration (AMD) and other retinal diseases are leading causes of vision impairment worldwide (1). Routine eye exams for older adults often include Optical Coherence Tomography (OCT) scans, which provide high-resolution images of retinal layers to detect signs such as swelling, abnormal growths, or thinning. However, interpreting OCT scans is labor-intensive and prone to variability (3).
This project aims to develop a deep learning model to classify OCT scans into six categories:
- Age-related Macular Degeneration (AMD)
- Diabetic Macular Edema (DME)
- Diabetic Retinopathy (DR)
- Normal (Normal)
Automated classification can accelerate screening, improve diagnostic accuracy, and reduce manual workload for ophthalmologists. Deep learning is particularly suitable because it can learn hierarchical features from high-dimensional OCT data, capturing subtle patterns that traditional methods may miss, and generalizing to new patients regardless of race or gender, or new imaging conditions.
