Medios-ai
Medios-ai has been studied in 2 papers since 2025: 2 observational studies in people; 2 in the past year. The newest paper, Jun 24, 2026: “Smartphone-based offline AI for multi-disease retinal screening: Real-world accuracy”.
2 papers study it5 findingsUpdated
At a glance
The findings reach people, from one lab or study. 2 independent labs in 1 country.
“For glaucoma (n = 109), sensitivity was 98.2% (95% CI: 94-100), specificity 99.0% (95% CI: 97-100), AUROC 0.99.”
None of its findings points the other way.
- 2 independent labs in 1 country
- Humans2
What did studies of medios-ai find?
1Studies of medios-ai measured one outcome, Misclassification of leukemic retinopathy as diabetic retinopathy.
Misclassification of leukemic retinopathy as diabetic retinopathy
upin humans1
1 study
Misclassification of leukemic retinopathy as diabetic retinopathy
upin humans1
Medios-ai is linked with misclassification of leukemic retinopathy as diabetic retinopathy in humans (patients with leukemic retinopathy) (eight patients incorrectly flagged).
“eight patients with leukemic retinopathy were incorrectly flagged as having DR, revealing a lack of specificity in distinguishing vascular retinal pathologies.”
Humanspatients with leukemic retinopathyretinan = 8
Cureus
What is medios-ai linked with?
4Studies link medios-ai with 4 measures, among them diabetic retinopathy, age-related macular degeneration and glaucoma.
Diabetic retinopathy
upin humans1
1 study
Diabetic retinopathy
upin humans1
Medios-ai predicts diabetic retinopathy in humans (DR screening) (AUROC 0.92, sensitivity 84.6%, specificity 99.0%).
“For DR (n = 78), sensitivity was 84.6% (95% CI: 75-92), specificity 99.0% (95% CI: 97-100), AUROC 0.92.”
Human trialDR screeningn = 78
Smartphone-based offline AI for multi-disease retinal screening: Real-world accuracy
European journal of ophthalmology
Age-related macular degeneration
upin humans1
1 study
Age-related macular degeneration
upin humans1
Medios-ai predicts age-related macular degeneration in humans (AMD screening) (AUROC 0.93, sensitivity 88.9%, specificity 97.5%).
“For AMD (n = 56), sensitivity was 88.9% (95% CI: 77-96), specificity 97.5% (95% CI: 95-99), AUROC 0.93.”
Human trialAMD screeningn = 56
Smartphone-based offline AI for multi-disease retinal screening: Real-world accuracy
European journal of ophthalmology
Glaucoma
upin humans1
1 study
Glaucoma
upin humans1
Medios-ai predicts glaucoma in humans (glaucoma screening) (AUROC 0.99, sensitivity 98.2%, specificity 99.0%).
“For glaucoma (n = 109), sensitivity was 98.2% (95% CI: 94-100), specificity 99.0% (95% CI: 97-100), AUROC 0.99.”
Human trialglaucoma screeningn = 109
Smartphone-based offline AI for multi-disease retinal screening: Real-world accuracy
European journal of ophthalmology
Detection of common retinal diseases
upin humans1
1 study
Detection of common retinal diseases
upin humans1
Medios-ai predicts detection of common retinal diseases in humans (oncology patients; comparison with clinical examination) (Glaucomatous cupping in three patients, DR in two cases, and signs of ARMD in two patients).
“The AI system accurately identified glaucomatous cupping in three patients, flagged two cases of DR, and detected signs of ARMD in two patients-all consistent with clinical examination.”
Humansoncology patients; comparison with clinical examinationretinan = 98
Cureus