Heidelberg Engineering, Inc. is collaborating with Eye2Gene to advance ophthalmic diagnostics and personalized treatment approaches for patients diagnosed with inherited retinal disease (IRD).
Let’s start with Heidelberg.
Founded in 1991 and based in Heidelberg, Baden-Württemberg, Germany, Heidelberg Engineering is a high-tech imaging solutions company that designs, manufactures, and distributes ophthalmic diagnostic devices.
- Note: This isn’t the company’s first partnership. Back in September 2023, Heidelberg and Orbis International collaborated on virtual vision service training and research funding to fight avoidable blindness. Read our coverage here.
And what exactly is Eye2Gene?
Based out of the University College London Institute of Ophthalmology, Eye2Gene is a research project funded by the National Institute for Health and Care Research (NIHR) and Moorfields Eye Charity.
The project is a collaboration between the following institutions and organizations:
- University College London
- Moorfields Eye Charity
- Oxford and Liverpool hospitals (United Kingdom)
- National Hospital Organization Tokyo Medical Center (Japan)
And what’s its purpose?
Essentially, Eye2Gene seeks to engineer artificial intelligence (AI) that can recognize and comprehend eye scans, and then use that information to predict the type of IRD a patient has.
Further: This AI would then specify which gene is likely to be affected as a result of that patient’s IRD.
How will this work?
Via a six-step process that includes:
- Applying to the NHS Health Research Authority to collect patient data
- Sourcing de-identified patient data and eye scans from four hospitals
- Eye care practitioners (ECPs) reviewing scans to confirm specific IRDs and eye damage signs in each patient
- Building the Eye2Gene system, which would include an AI algorithm using the ECPs’ labeled scans to recognize IRD damage and determine which disease it was caused by
- Eye2Gene system copying human ECPs to look at brand new, unlabeled eye scans and predict which IRD is shown
- Human ECPs will inform the system if it’s correct; the process will repeat until the system is deemed to be reliably identifying IRDs
- Sending the Eye2Gene system to UK regulator for medical device approval; if successful, the technology could be implemented in the healthcare system
Gotcha. So how does Heidelberg come into play?
The company’s image quality and workflow solutions will combine with the Eye2Gene AI algorithm.
Specifically: The Eye2Gene algorithm will be integrated with the following Heidelberg technologies:
- SPECTRALIS BluePeak Autofluorescence
- Non-invasive scanning laser fundus imaging modality that shows metabolic stress within the retina via lipofuscin (used as an indicator)
- Reveals retinal pigment epithelial (RPE) and photoreceptor cell malfunctions
- Confocal scanning laser ophthalmoscopy (cSLO) Infrared and optical coherence tomography (OCT) images
- HEYEX 1’s Heidelberg Appway
- Heidelberg Eye Explorer (HEYEX) 1: image management platform
- Streamlines and provides simple access to diagnostic images, reports, and maps; enables image acquisition, management, and review within one platform
- Connects with all Heidelberg instruments
- Heidelberg Appway: a novel gateway solution for secure data transfer via cloud exchange with external ophthalmic application providers
- Enables access to clinical and research applications for diagnostic image analysis
- Operates within the HEYEX 1 software
- Enables access to clinical and research applications for diagnostic image analysis
- Heidelberg Eye Explorer (HEYEX) 1: image management platform
And when combined?
The intent is for researchers to determine “how images acquired with the SPECTRALIS can be interrogated by Eye2Gene to predict which gene is causing the inherited disease,” according to the companies.
The significance of this: A potential glimpse of what the future of precision medicine in ophthalmology may look like, Heidelberg stated.
Go on …
According to Nikolas Pontikos, Eye2Gene’s lead researcher, by training and validating a deep learning network with SPECTRALIS data (de-identified and obtained via Eye2Gene’s collaborators), “we can engineer an expert AI solution that may play an important role in democratizing expert knowledge for (IRDs).”
The benefit of this partnership? Such a collaboration of technologies may not only validate Eye2Gene’s science but also provide the project a platform to transition its technology into future clinical practices, Pontikos noted.
So where can I see this integration in action?
ECPs can try a demo of the Eye2Gene app by clicking here.
And if you’re attending the 2024 Association for Research in Vision and Ophthalmology (ARVO) annual meeting, you’re in luck! Attendees can visit Booth #1905 to see a live demonstration of the HEYEX2 with the Heidelberg Appway—plus learn more about retinal scan integration for IRD prediction
Lastly, any ARVO research on this topic?
There is! The Eye2Gene research project team is presenting two poster presentations:
- B0397: Creating the World’s Largest Dataset of Segmented Inherited Retinal Disease Features by Bootstrapping Manual Annotations with AI
- Presented by William Woff
- When: Tuesday, May 7 (1:15 to 3 pm PST)
- Where: Exhibit Hall - Seattle Convention Center
- B0111: Extending Eye2Gene to Quantify the Phenotypic Diversity and Similarity of 63 Inherited Retinal Diseases Using an Embedding Approach
- Presented by: Bernardo Souza Mendes
- When: Wednesday, May 8 (10:30 am to 12:15 pm PT)
- Where: Exhibit Hall - Seattle Convention Center