DeepRetina: Alzheimer’s Disease Detection Using AI

Authors:

Mr. Vishva Kiran R C¹, Kavya Shree R²_, Likhitha H³, Manaswini Anand M⁴, Nettem Chandana⁵

Assistant Professor, Dept of Computer Science and Engineering, KSIT, Karnataka, India¹

Student, Dept of Computer Science and Engineering, KSIT, Karnataka, India^2-5

Abstract:

Alzheimer’s Disease (AD) is a progressive neurodegenerative disorder characterized by cognitive decline and memory loss. Early diagnosis remains a major challenge due to the invasive, expensive, and limited accessibility of conventional diagnostic methods such as PET scans and cerebrospinal fluid analysis.

Recent research highlights the retina as a promising non-invasive biomarker for detecting neurodegeneration, owing to its direct connection with the brain. Advanced imaging techniques such as Optical Coherence Tomography (OCT) and Optical Coherence Tomography Angiography (OCTA) enable detailed visualization of retinal layers and microvasculature.

This survey reviews current approaches that use retinal imaging combined with Artificial Intelligence (AI), particularly deep learning models, to detect Alzheimer’s Disease. Key indicators such as Retinal Nerve Fiber Layer (RNFL) thinning, Ganglion Cell Layer (GCL) degeneration, and amyloid-beta plaque accumulation are analyzed.

This survey demonstrates that AI-integrated retinal imaging frameworks, such as DeepRetina, offer a scalable paradigm for early Alzheimer’s Disease detection. Deep learning models, particularly CNNs, enable precise quantification of retinal structural and microvascular biomarkers from OCT/OCTA data. These approaches demonstrate high sensitivity and specificity while remaining non-invasive and cost-effective. Further advances in model generalization and large-scale clinical validation are essential for real-world deployment.

Keywords:

Alzheimer’s Disease, Retinal Imaging, Optical Coherence Tomography (OCT), Deep Learning, Convolutional Neural Networks (CNN), Retinal Nerve Fiber Layer (RNFL), Early Detection.