Scientists at the National Eye Institute (NEI), part of the National Institutes of Health (NIH) in the United States, have created a detailed digital model of human eye cells to improve research into age-related macular degeneration (AMD). The work was reported in February 2026 and published in a peer-reviewed journal.1
This advancement provides researchers with a new platform to examine cellular changes linked to AMD, which is a major cause of central vision loss in older adults worldwide.
Age-related macular degeneration is a progressive eye condition that affects the retina’s central region (macula). In AMD, supporting cells beneath light-sensing cells deteriorate, leading to distortion or loss of central vision. The disease typically presents in adults over 50 and can impact daily tasks such as reading and facial recognition.
The NIH team focused on retinal pigment epithelial (RPE) cells, which are essential for maintaining photoreceptors. These cells exhibit polarity, meaning they have distinct structural orientation from top (apical) to bottom (basal). Loss of polarity is a hallmark of early AMD.
To map RPE structure, researchers gathered more than 1.3 million high-resolution images of these cells using automated confocal microscopy. They trained an artificial intelligence system, named POLARIS, to analyze and quantify features such as nuclei, organelles, and cell shape.
The result is a three-dimensional, data-driven digital twin representing the organization of healthy RPE cells. This model can serve as a comparative reference for diseased states.
As published in the NIH news release, Kapil Bharti, PhD, senior investigator at the National Eye Institute (part of the National Institutes of Health), explained the significance of the digital twin model in understanding retinal disease.2
This work represents the first ever subcellular resolution digital twin of a differentiated human primary cell, demonstrating how the eye is an ideal proving ground for developing methods that could be used more generally in biomedical research.
Kapil Bharti, Ph.D., Scientific Director, NIH’s National Eye Institute (NEI)
The AI model learns from large datasets of images to construct a virtual representation of RPE cell architecture. It identifies patterns of organization and structural changes that occur during cell development or disease progression. Researchers can use this digital twin to examine how RPE cells behave in normal conditions compared with disease states.
Davide Ortolan, Ph.D., the study’s first author and an NEI research fellow, said the 3D digital map helped identify cell state transitions involved in apical-basal polarity.2
By combining AI with mathematical modeling, we've created a window into cellular processes that were previously hidden from view. This technology doesn't just help us understand what's happening in AMD, it gives us a platform to discover how to fix it.
Davide Ortolan, Ph.D., NEI Research Fellow
The digital twin is not a clinical diagnostic tool, but it provides a research benchmark for understanding early cellular changes in AMD. By comparing normal and abnormal cell organization, scientists can investigate:
How cell polarity is disrupted early in AMD
Structural changes that precede vision loss
Mechanisms that could be targeted for therapeutic development
This model could accelerate research aimed at discovering interventions before irreversible damage occurs.1
Beyond AMD, this approach can be applied to other diseases where cell organization changes are important. Digital replicas based on large imaging datasets and AI analysis may support research in neurodegenerative diseases, cancer biology, and tissue development.
The digital twin approach represents a powerful new tool for AMD therapeutic development and could be adapted to study other eye and non-eye diseases and conditions affecting cell polarity.
Kapil Bharti, Ph.D., Scientific Director, NIH’s National Eye Institute (NEI)
The development of a high-resolution digital twin of RPE cells marks a new research capability in AMD biology. By combining AI, imaging, and mathematical modeling, scientists now have a tool to explore the subcellular organization of eye cells with precision. While the model itself does not treat AMD, it strengthens the foundation for future discoveries that may one day improve prevention and therapy.
1. Ortolan, D., P. Sathe, A. Volkov, et al. 2026. “AI Driven 3D Subcellular RPE Map Discovers Cell State Transitions in Establishment of Apical-Basal Polarity.” npj Artificial Intelligence 2: 20. https://doi.org/10.1038/s44387-026-00074-6.
2. National Institutes of Health. 2026. “NIH Scientists Develop ‘Digital Twin’ of Eye Cells to Understand and Treat Age-Related Macular Degeneration.” National Institutes of Health News Release, February 10, 2026. https://www.nih.gov/news-events/news-releases/nih-scientists-develop-digital-twin-eye-cells-understand-treat-age-related-macular-degeneration.
(Rh/SS/MSM)
