EUVIMEDEUVIMED entdecken
Crossref

P-161: Automated diabetic retinopathy detection in smartphone based fundus photography using artificial intelligence conducted at health camp in Assam

P-161: Automatisierte Erkennung diabetischer Retinopathie in smartphonebasierter Fundusfotografie mit künstlicher Intelligenz in einem Gesundheitscamp in Assam

A Raha

Hormone Research in Paediatrics · 04.11.2026 · 99 (Supplement_4)

DOI: 10.1159/hrp/adaag020.220 ↗

Auf einen Blick

Bei 501 Patient:innen mit Typ-2-Diabetes erkannte die KI diabetische Retinopathie und STDR mit hoher Sensitivität. Die Spezifität lag bei rund 80 %.

Fragestellung

Wie gut erkennt eine KI-basierte Software diabetische Retinopathie und sight-threatening DR anhand von Smartphone-Fundusfotografie im Vergleich zur ophthalmologischen Beurteilung?

Methode

1775 Patient:innen mit Typ-2-Diabetes erhielten Fundusfotografie mit Remidio Fundus on phone bei einem Gesundheitscamp an 5 Orten in Assam; 501 Bilder wurden beurteilt. Vergleich der KI-Software EyeArtTM mit Ophthalmolog:innen nach ICDR.

Ergebnisse

DR: Ophthalmolog:innen 323 (64,4 %) vs. KI 343 (68,6 %); Sensitivität 95,8 % (95 %-KI 92,9–98,7), Spezifität 80,2 % (95 %-KI 72,6–87,8). STDR: 189 (37,8 %) vs. 246 (49,3 %); Sensitivität 99,1 % (95 %-KI 95,1–99,9), Spezifität 80,4 % (95 %-KI 73,9–85,9).

Schlussfolgerung der Autor:innen

Die automatisierte KI-Analyse von FOP-Smartphone-Retinabildern hat eine sehr hohe Sensitivität für DR und STDR und kann als initiales Screening-Tool für die Massenuntersuchung dienen.

Kernaussagen

  • DR-Erkennung: Sensitivität 95,8 %, Spezifität 80,2 %
  • STDR-Erkennung: Sensitivität 99,1 %, Spezifität 80,4 %
  • Vergleich mit ophthalmologischer ICDR-Grading-Bewertung

Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.

Kurz gefasst

Fachartikel in der Zeitschrift Hormone Research in Paediatrics, online erschienen am 04.11.2026. Verfasst von A Raha.

Auf einen BlickAbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

1 in 3 diabetics is at risk of developing diabetic retinopathy (DR). 1 in 10 Diabetics suffers vision loss. 80% of all vision impairment globally is avoidable.

Objectives

The study was conducted to assess the role of artificial intelligence (AI)-based automated software for detection of diabetic retinopathy (DR) and sight-threatening DR (STDR) by fundus photography taken using a smartphone-based device and validate it against ophthalmologist’s grading.

Methods

1775 patients with type 2 diabetes underwent retinal photography with Remidio ‘Fundus on phone’ (FOP), a smartphone-based device, at health camp conducted at 5 places in the periphery of Lumding constituency of Assam. Out of 1775 patients, 501 patients were found to be having Diabetic Retinopathy of various grade. Grading of DR was performed by the ophthalmologists using International Clinical DR (ICDR) classification scale. STDR was defined by the presence of severe non-proliferative DR, proliferative DR or diabetic macular oedema (DME). The retinal photographs were graded using a validated AI DR screening software (EyeArtTM) designed to identify DR, referable DR (moderate non-proliferative DR or worse and/or DME) or STDR. The sensitivity and specificity of automated grading were assessed and validated against the ophthalmologists’ grading.

Results

Retinal images of 501 patients were graded. DR was detected by the ophthalmologists in 323 (64.4%) and by the AI software in 343 (68.6%) patients while STDR was detected in 189 (37.8%) and 246 (49.3%) patients, respectively. The AI software showed 95.8% (95% CI 92.9–98.7) sensitivity and 80.2% (95% CI 72.6–87.8) specificity for detecting any DR and 99.1% (95% CI 95.1–99.9) sensitivity and 80.4% (95% CI 73.9–85.9) specificity in detecting STDR.

Conclusions

Automated AI analysis of FOP smartphone retinal imaging has very high sensitivity for detecting DR and STDR and thus can be an initial tool for mass retinal screening in people with diabetes for early detection and timely referral to ophthalmologist.

Zitieren

APA

Raha, A. (2026). P-161: Automated diabetic retinopathy detection in smartphone based fundus photography using artificial intelligence conducted at health camp in Assam. Hormone Research in Paediatrics, 99(Supplement_4). https://doi.org/10.1159/hrp/adaag020.220

Vancouver

Raha A. P-161: Automated diabetic retinopathy detection in smartphone based fundus photography using artificial intelligence conducted at health camp in Assam. Hormone Research in Paediatrics. 2026;99(Supplement_4). doi:10.1159/hrp/adaag020.220

BibTeX
@article{Raha2026,
  author = {Raha, A},
  title = {P-161: Automated diabetic retinopathy detection in smartphone based fundus photography using artificial intelligence conducted at health camp in Assam},
  journal = {Hormone Research in Paediatrics},
  year = {2026},
  volume = {99},
  number = {Supplement_4},
  doi = {10.1159/hrp/adaag020.220}
}
Kostenlos starten →