AI and Model Refinement: New Tools Aim to Accelerate Eye Disease Cures

DENVER, CO – April 09, 2026 – The development of new treatments for debilitating eye conditions like wet age-related macular degeneration (wet-AMD) and diabetic retinopathy faces a significant hurdle: the long, complex, and costly preclinical research phase. Now, new advancements in artificial intelligence and a deeper understanding of biological models are poised to break down these barriers. Next month, preclinical research specialist Powered Research is set to present two pivotal studies at the Association for Research in Vision and Ophthalmology (ARVO) 2026 conference, the world's premier gathering of eye and vision scientists.

The presentations detail an innovative AI-driven tool designed to rapidly assess treatments for retinal diseases and critical findings on animal model selection for corneal pain studies. Together, they represent a significant step forward in improving the speed, accuracy, and translatability of the foundational science that underpins the next generation of ophthalmic therapies.

An AI Co-Pilot for Retinal Drug Discovery

The first breakthrough to be presented addresses a core challenge in treating neovascular diseases like wet-AMD. These conditions are characterized by the growth of abnormal, leaky blood vessels in the retina, a process called neovascularization. In preclinical studies, evaluating the effectiveness of a potential drug requires precisely measuring its ability to reduce this leakage and shrink the problematic vessels.

Traditionally, this analysis is a labor-intensive process, often relying on manual or semi-automated methods that can be time-consuming and prone to subjective variability. Powered Research's solution is a machine learning-based image analysis tool designed to bring unprecedented speed and objectivity to this process. The tool, which will be detailed in a poster session titled "Artificial Intelligence to Quantify Leakage of Neovascular Tufts from Fundus Imaging," automates the analysis of fluorescein angiography (FA) images, a standard diagnostic technique where a fluorescent dye is used to visualize blood flow and leakage in the retina.

The study, led by a team including Zsolt Ablonczy and Daniel Costa, utilized a preclinical rabbit model that mimics the vascular leakage seen in human patients. The AI tool, a trained pixel classifier, was used to quantify the total vascular area from these images. To validate its accuracy, the results were rigorously compared against two established methods: a traditional fluorescence assessment in ImageJ and a separate AI-assisted analysis of retinal tissue in vitro.

The results were compelling. In models treated with aflibercept, a standard-of-care therapy, the AI tool measured an average decrease in total vascular area of 40% at two weeks and 41% at four weeks. These findings were consistent with data from the other two quantification methods, confirming the tool's reliability. The conclusion is clear: the new AI-assisted analysis provides a rapid, unbiased, and noninvasive method for evaluating the impact of novel pharmaceuticals.

For pharmaceutical companies, the implications are profound. A high-throughput tool like this can dramatically accelerate the screening of potential drug candidates, allowing researchers to get more reliable data more quickly. This efficiency can reduce R&D costs and, most importantly, shorten the timeline for getting promising new therapies into clinical trials and eventually to the patients who need them.

Precision in Practice: Refining the Foundations of Corneal Research

While sophisticated software is revolutionizing data analysis, the integrity of that data begins with the biological model itself. Powered Research's second presentation at ARVO 2026 highlights this often-overlooked but critical aspect of preclinical science. This study delves into the subtle but significant differences between two rabbit strains commonly used in ophthalmology research: the New Zealand White (albino) and the Dutch Belted (pigmented).

Rabbits are frequently chosen for eye research because the size and structure of their eyes are more comparable to humans than those of smaller rodents, making the results more translatable. However, the study, titled "Correlation Between Nerve Density and Corneal in Two Rabbit Strains," reveals that not all rabbits are created equal when it comes to modeling conditions related to the cornea.

The research team, which included Flavia Leao Barbosa and veteran veterinary ophthalmologist Brian Gilger, assessed central corneal sensitivity and nerve density in 57 rabbits across the two strains. Corneal nerves are essential for maintaining a healthy ocular surface and are central to conditions involving corneal pain and neuropathy.

The findings demonstrated a stark difference. Dutch Belted rabbits had significantly lower corneal sensitivity compared to their New Zealand White counterparts. Using Cochet-Bonnet esthesiometry, a device that measures sensory thresholds, the team recorded a mean sensitivity nearly twice as high in the albino rabbits. This functional difference was then linked to a physical one: microscopic analysis revealed that while pigmented rabbits had denser superficial nerve projections, the albino rabbits had a denser network of sub-basal nerves, which are closely linked to sensory perception.

This discovery is more than an academic curiosity; it has direct practical implications. It emphasizes that the choice of animal strain is a critical variable that can profoundly influence the outcome of studies focused on corneal pain, nerve regeneration, or treatments for ocular surface disease. Using a less sensitive strain to test a new analgesic, for example, could lead to misleading conclusions about the drug's efficacy. These findings provide a crucial roadmap for scientists, enabling them to select the appropriate model for their specific research question, thereby improving the accuracy and reliability of their data.

A Unified Strategy for Better, Faster Therapies

Viewed together, the two abstracts from Powered Research showcase a comprehensive strategy for enhancing preclinical ophthalmology research. One innovation provides a powerful new tool for analysis, while the other provides a more refined understanding of the model being analyzed. This dual focus on both technology and fundamental biology is essential for building a more robust and efficient drug development pipeline.

By presenting these findings at ARVO, Powered Research is sharing its advancements on the most prestigious stage in vision science, inviting scrutiny and collaboration from thousands of leading clinicians, researchers, and industry partners. The credibility of the research is further bolstered by the multidisciplinary expertise of the author teams, which combine skills in AI, pharmacology, and comparative ocular pathology.

Ultimately, this work is about bridging the gap between the laboratory and the clinic. By creating faster, more accurate analytical tools and ensuring the use of the most appropriate biological models, researchers can generate higher-quality data. This, in turn, increases the likelihood that drug candidates succeeding in preclinical trials will also succeed in humans. This foundational work, though steps removed from the pharmacy shelf, is essential for building the pipeline of future treatments that could one day preserve sight for millions.