AgingResearch.NewsJoin free
Ophthalmology Science

German cohort offers a resource for eye health research

Eye assessments from 48,460 adults aged 19–74 provided a population-based resource that could support artificial intelligence research.

Paper by Camila Roa, Ansgar Beuse, Alexandra Schweig, +23 more authorsPhilipp Berens · Tübingen AI Center
Figure 1. Flowchart of study participants for the baseline examination in the level 2 ophthalmology submodule of the NAKO.
Open the figure at full size
Figure 1. Flowchart of study participants for the baseline examination in the level 2 ophthalmology submodule of the NAKO.Flowchart of study participants for the baseline examination in the level 2 ophthalmology submodule of the NAKO.Roa et al.

Our read

AI & data
Studied:
Cohort study of 48,460 participants undergoing ophthalmic assessment, against UKB ophthalmic cohort (Multicenter)
What changed:
High-quality retinal images were more common in NAKO than UKB: 68.2% against 48.7%
The catch:
Only 40.8% of images were gradable in the oldest age group
Why it’s aging research:
It builds or tests a tool for aging research.

Editor of record: Boris Djordjevic · AgingResearch.News

Paper
Follow @agingbrief

Researchers analyzed baseline eye examinations from 48,460 adults aged 19–74 in the German National Cohort, an observational study. Participants had an average age of 48.9 years, and 52.7% were male. The assessments included tests of vision sharpness, color photographs of the back of the eye and a brief eye health history. Researchers also collected general health measurements and questionnaire responses.

Four artificial intelligence models classified 68.2% of the eye images as suitable for assessment. Measurements of eye structures differed systematically across age groups. The researchers also trained artificial intelligence models to predict age, sex and blood pressure from the images. They concluded that the data could support the development and testing of artificial intelligence tools for eye health research.

Why it matters

Combining eye images with age and general health measurements could help researchers investigate how eye health relates to aging.

Caveats

This baseline observational analysis compared age groups rather than tracking how individuals’ eyes changed over time. The findings concern adults aged 19–74 in Germany.

From the paper

Participants

The data set consisted of 48 460 participants who had taken part in the baseline ophthalmology module and had ≥1 fundus image for ≥1 eye and a visual acuity examination, as provided by the NAKO TransferHub.

Results, Population Profile

Limitation

However, its large-scale application in population-based cohorts remains limited.

Strengths and Limitations

Quoted word for word from the full text on PubMed Central.

Comments

Be the first to comment.

More on Deep learning