- 200,000+ patients: AI model analyzes over a decade of clinical records from de-identified patients.
- 19.8 million Americans: Estimated number living with age-related macular degeneration (AMD).
- 9.6 million people: Affected by diabetic retinopathy, with nearly 2 million facing vision-threatening forms.
Experts would likely conclude that this AI-driven data extraction represents a significant advancement in ophthalmology, enabling more precise research and personalized treatments for chronic eye diseases.
AI Unlocks a Decade of Eye Data to Fight Leading Causes of Blindness
NEW YORK, NY – September 15, 2026 – In the data-rich field of ophthalmology, a critical blind spot has persisted for decades. The most vital information on how chronic eye diseases progress—subtle changes in retinal structure, responses to treatment, and measurements tracked over years—has remained trapped in the digital equivalent of filing cabinets: scanned reports, PDF attachments, and unstructured physician notes. Today, a landmark partnership aims to bring that data into the light.
Century Health, a firm specializing in clinical AI, has joined forces with Eye Health America (EHA), a major integrated eye care platform in the Southeastern U.S. Their goal is to deploy an advanced AI model to parse over a decade of clinical records from more than 200,000 de-identified patients. The collaboration will create one of the most detailed real-world datasets ever assembled for chronic retinal diseases, targeting two of the leading causes of vision loss in America: diabetic retinopathy and age-related macular degeneration (AMD).
The Digital Blind Spot in Medicine
Ophthalmology is a specialty of millimeters and microns, where treatment decisions hinge on tracking minute changes over long periods. Yet, the standard electronic health record (EHR) systems used across medicine are notoriously poor at capturing this longitudinal, narrative data in a structured, searchable format. While structured fields for diagnosis codes or prescriptions are easily analyzed, the crucial context—the why behind a clinical decision—is often buried in free-text notes or image reports.
This “unstructured data” problem means that large-scale research has been operating with an incomplete picture. Studies relying solely on EHR fields miss the granular progression metrics that clinicians meticulously record. As a result, drug developers and researchers have a limited view of how diseases like AMD and diabetic retinopathy truly evolve in the real world and how diverse patient populations respond to therapies outside the rigid confines of a clinical trial.
This partnership directly confronts that challenge. Century Health's AI platform, known as the Century Health Abstraction and Retrieval Model (CHARM), is designed to read and interpret these complex, unstructured documents. By applying sophisticated natural language processing, CHARM can identify and extract key quantitative values and disease progression metrics, effectively translating years of clinical history into research-grade, structured data. This process unlocks a trove of information that was previously too labor-intensive and costly to access at scale.
A New Weapon Against Chronic Vision Loss
The public health stakes are immense. An estimated 19.8 million Americans live with AMD, a number that rises sharply with age. Meanwhile, diabetic retinopathy, a complication of diabetes, affects nearly 9.6 million people, with almost 2 million facing a vision-threatening form of the disease. For these patients, treatment is not a one-time event but a long-term journey of monitoring and intervention to preserve sight.
"Caring for patients with chronic eye conditions means following subtle changes over many years, yet much of that longitudinal history isn't available for large-scale research," said Dr. Cathleen McCabe, Chief Medical Officer at Eye Health America. "By transforming more clinical history into usable data, we can better understand how these diseases progress, identify which treatments work best for different patients, and ultimately improve the care we deliver."
The dataset created through this initiative will provide an unprecedented window into these disease trajectories. Researchers will be able to analyze how different patient subgroups respond to treatments, identify early predictors of rapid progression, and discover patterns that are invisible in smaller, less detailed studies. This concept, known as Real-World Evidence (RWE), is becoming a cornerstone of modern medical research, offering insights that complement the findings of traditional randomized controlled trials (RCTs) by reflecting the complexities of routine care.
The Strategic Blueprint for Data-Driven Healthcare
This alliance represents more than just a technological breakthrough; it is a strategic blueprint for the future of healthcare data. For Eye Health America, a sprawling network of ophthalmology practices and surgery centers, the partnership unlocks the immense latent value of its clinical data archives. It transforms a byproduct of patient care into a powerful asset for advancing medical science, enhancing EHA's reputation as an innovator in the field.
For Century Health, the collaboration is a powerful validation of its technology. Processing a decade's worth of complex data from a major provider like EHA demonstrates the scalability and real-world applicability of its CHARM platform. It positions the company as a key enabler in the burgeoning market for AI-driven clinical data extraction, a sector attracting significant investment as pharmaceutical companies and research institutions race to leverage RWE.
"The clinical data that matters most in retinal disease, like progression measurements, imaging findings, treatment response over years, has remained locked inside ophthalmology practices," noted Vish Srivastava, Co-Founder and CEO of Century Health. "We're thrilled to partner with Eye Health America to change that, and to bring this depth of real-world evidence to life sciences researchers."
Of course, handling such a vast repository of patient information requires an unwavering commitment to privacy. The entire process hinges on robust de-identification protocols that strip the records of all 18 personal identifiers stipulated by the Health Insurance Portability and Accountability Act (HIPAA), ensuring that patient privacy is rigorously protected while the scientific value of the data is preserved. By converting clinical histories into anonymized, aggregated insights, the partnership navigates the critical intersection of innovation and ethics.
As this newly structured data becomes available to the life sciences community, it promises to accelerate the development of more personalized and effective therapies. By finally illuminating the details of long-term disease progression, this visionary alliance may help create a future where fewer people lose their sight to these devastating chronic conditions.
Topics & Related
📝 This article is still being updated
Are you a relevant expert who could contribute your opinion or insights to this article? We'd love to hear from you. We will give you full credit for your contribution.
Contribute Your Expertise →