Post

New research · Ophthalmology
Cureus · 5d
AI / informaticsCureus · 2026

Mixing Synthetic and Real Images Improves Artificial Intelligence-Based Detection of the Pupil, Iris, and Sclera: A Cross-Domain Validation Study.

Krishna Keshav, Deepsekhar Das, Sumit Grover … Avinav Bharti
Read paper
OphthalmologyAI / informatics

Mixing synthetic and real eye images gives artificial intelligence stable, accurate pupil detection.

Mixing Synthetic and Real Images Improves Artificial Intelligence-Based Detection of the Pupil, Iris, and Sclera: A Cross-Domain Validation Study.

Krishna Keshav … Avinav Bharti
Cureus · 2026
Background

This study aimed to evaluate whether mixing synthetic and real eye images for artificial intelligence (artificial intelligence) training improves cross-domain segmentation of the sclera, iris, and pupil compared to single-domain datasets.

Purpose

This study aimed to evaluate whether mixing synthetic and real eye images for artificial intelligence (artificial intelligence) training improves cross-domain segmentation of the sclera, iris, and pupil compared to single-domain datasets.

Methods

Four Roboflow 3.0 instance segmentation models were trained: (1) 100 artificial intelligence-generated images, (2) 100 real images, (3) 50:50 mixed, n = 100, and (4) 50:50 mixed, n = 200.

n = 200 images
Results
88.5%
pupil detected correctly in nearly 9 of 10 test images
n = 200 images
More results

Mixed models eliminated domain-specific failures.

Pupil accuracy showed a significant training × test domain interaction (p = 0.003), with single-domain models failing on opposite domains: artificial intelligence-trained 0% on one real image; real-trained 50% on one artificial intelligence image.

More results

Doubling mixed data to 200 images gave no added benefit (p = 0.95).

“
Conclusion

Hybrid training with 50:50 synthetic-real images achieves robust, domain-stable detection of the pupil, iris, and sclera. Mixed datasets, not larger datasets, are key for clinically deployable ocular artificial intelligence.

Read paper
0 comments

No comments yet. Be the first.

Related papers

LatestFoundational
AI / Informatics
0·962 for OCT
MerMED-FM was very accurate at diagnosing diseases using eye scans
AI / Informatics
0.98
Artificial intelligence's eyelid-height measurements matched doctors' manual measurements almost perfectly
AI / Informatics
92.1%
combining eye scans and photos correctly told benign from cancerous lesions apart nearly every time
Cohort Study
28.6%
recurred locally in more than 1 in 4 patients over years of follow-up
Cohort Study
-0.395
excision group's post-op eyelid fullness score was lower - greater improvement
Cohort Study
86.7%
of infants probed after 12 months still had unresolved tear duct blockage
Observational
16%
eyelid tissue in rosacea patients showed less of this key repair-signaling protein inside cell nuclei
Cohort Study
25%
about 1 in 4 treated cases had symptoms return after improving