Skip to content

Eyes on the Prize: The Case for AI in Ophthalmic Clinical Research

Eyes on the Prize: The Case for AI in Ophthalmic Clinical Research

Gettyimages 864001400 Web

The pace of innovation in ophthalmic research continues to advance, yet costly issues with clinical trial recruitment, management and analysis persist. Instead of settling for the status quo, perhaps now is the time to revisit the potential of Artificial Intelligence (AI) as a means to overcome the bottlenecks and inefficiencies that plague the ophthalmology sector?

In this blog we’ll dive deeper into the emerging benefits of AI assisted data management and analysis – with a special emphasis on ophthalmic clinical trials. Given the average length of a clinical trial from start to market is currently 7.5 years, efficiencies are welcome, and AI clearly has the potential to significantly accelerate ophthalmic research and resulting treatments and outcomes in several key areas:

Image Grading Accuracy: Manually grading images is excessively time consuming and is prone to human error and notoriously variable across readers. And when manual grading processes are used for large quantities of ophthalmic image data, errors only increase, and clinical trials can drag out extensively as the time to results drags at the reading centers.

Recruitment and Retention of Participants: The longer time frames of clinical trials create challenges not only in recruiting but also in retaining participants. 40% of clinical trials in ophthalmology are discontinued due to poor recruitment. With AI, trial managers can analyze all the data available for a participant to proactively determine if they would be a good fit for the trial.

Better Targeting of Subjects: Some trials take longer simply because certain diseases can vary among subjects in their time to progress. Imagine a trial for a drug that can prevent disease progression – in order to know definitively whether a subject is being helped by a drug or is just a slow progressor, trials need to recruit a large number of subjects and last a long time. AI-based prognostic models hold the promise to be used to “enrich” the trial population with fast-progressors – effectively lowering both the number of subjects required and the length of trial.

Data Capacity: For ophthalmic trials in particular, the number and size of the images and data grows exponentially over the trial period. Researchers are hard pressed to ensure the ability to store this massive data, as well as organize and index it to ensure efficient retrieval, analysis, and reporting. AI-enhanced workflows can streamline this process, saving time and effort.

Reproducible Results: An essential component of a successful clinical trial is generating reproducible results. Since AI-powered analysis uses a consistent approach when grading image data, the inherently subjective interpretation of the data can be factored out.

Cost: Without AI-driven processes, the aforementioned issues combine to increase clinical trial costs and overrun already stretched budgets. Ophthalmology has the second highest average per study costs across Phase 1, 2 & 3 trials. By applying AI to address manpower and analysis inefficiencies, clinical trials have a much higher chance to be completed on time and within budget.

What’s Next for AI in Ophthalmology

By reducing the time and effort required for clinical trials, AI-backed workflows can accelerate clinical trials, new treatments, and better patient outcomes.

To learn how Orion enables AI-backed analysis, provides interoperability in ophthalmic image analysis, and reduces read times with automated measurements, visit or get in touch to request a demo.

Cybersight Lecture: Artificial Intelligence in Ophthalmology

Further Reading:
Project Pro: Why data preparation is an important part of data science?