Scientists Develop Artificial Retina Phantom for Standardized Eye Disease Diagnosis
Scientists at the Korea Research Institute of Standards and Science (KRISS) have made a groundbreaking advancement in the field of ophthalmology. They have created an artificial retina phantom that faithfully replicates the intricate structural layers and microvascular network of the human retina. This innovation is set to revolutionize the way we evaluate and calibrate ophthalmic imaging devices, leading to more accurate and reliable diagnosis of retinal diseases.
The retina, often likened to a camera film, plays a crucial role in detecting light and transmitting visual information to the brain. With the increasing prevalence of retinal diseases due to factors like aging, excessive screen time, and genetic predisposition, early diagnosis and continuous monitoring are vital to prevent vision loss. However, the lack of a standardized reference for evaluating and calibrating diagnostic instruments has been a significant challenge in ensuring consistent and reliable diagnostic outcomes.
To address this issue, the Nanobio Measurement Group and the Medical Metrology Group at KRISS developed an artificial eye, the retina-mimicking eye phantom. This phantom is designed with precision, accurately reproducing the structure and function of the human retina. Its unique design, resembling a ruler with marked scales, enables accurate assessment of diagnostic device performance. When inserted into ophthalmic imaging systems, it provides objective verification and calibration of critical aspects such as image resolution and field of view.
The research team's achievement is particularly notable as it surpasses conventional retinal phantoms. While traditional phantoms only replicated a simplified version of the retinal vasculature, the KRISS phantom precisely reproduces all 13 structural layers of the retina, including its curvature, microvascular networks with fluidic flow, and retinal autofluorescence. With an impressive match rate of over 90% with a real human retina, this phantom can be applied to a wide range of diagnostic platforms, from tomographic imaging systems to angiography devices.
This development is expected to set a new benchmark for the standardization of medical imaging devices, enhancing the accuracy of retinal disease diagnosis and treatment monitoring. By providing a clear reference for evaluating and calibrating diagnostic instruments, medical institutions can ensure consistent and reliable test results for patients, regardless of the location of their retinal examinations.
The newly developed phantom is also anticipated to find widespread use in both industry and education. Manufacturers of retinal imaging devices can utilize the phantom to evaluate and refine prototype performance, ensuring consistent product quality during production. Additionally, medical professionals can leverage the phantom's realistic mimicry of the human retina for clinical training and diagnostic education, further enhancing their expertise.
Lee Sang Won, Head of the Nanobio Measurement Group at KRISS, emphasized the potential of AI-assisted diagnostic methods in the growing demand for retinal disease diagnosis. He stated, 'By calibrating ophthalmic imaging systems using this phantom, we can obtain high-quality training data, which will contribute to improving the performance of AI-based diagnostic devices.'
The research findings have been published in the journal Communications Engineering and are available online. The DOI for the publication is 10.1038/s44172-025-00475-6.
