Künstliche-Intelligenz-Systeme zur Erkennung der diabetischen Retinopathie

Zusammenfassung

Früherkennung und rechtzeitige Behandlung können in über 90% der diabetischen Retinopathie einen schweren Sehverlust verhindern. Die deutsche Nationale Versorgungsleitlinie empfiehlt, dass Personen mit Diabetes jährliche bzw. 2-jährliche Augenuntersuchungen zur Erkennung einer behandlungsbedürftigen diabetischen Retinopathie erhalten. Um den Herausforderungen gerecht zu werden, wurden KI-Algorithmen entwickelt, um DR autonom aus Fundusfotografien ohne menschliche Beurteilung zu erkennen. In den letzten Jahren haben viele KI-Algorithmen eine gute Sensitivität und Spezifität für die Erkennung einer behandlungsbedürftigen DR im Vergleich zu menschlichen Beurteilern erzielt. Bislang kommt ein KI-basiertes DR-Screening auch in Ländern mit besser entwickelter digitaler Infrastruktur als Deutschland nur in geringem Umfang zum Einsatz, denn viele Fragen wie Akzeptanz, Kosteneffektivität, Haftungsrisiken, IT-Sicherheit und die Kostenerstattung sind noch unzureichend beantwortet. In dieser Übersicht über KI-Anwendungen zum DR-Screening werden wichtige Konzepte in der Entwicklung und aktuell zugelassene KI-Algorithmen vorgestellt, die im Vergleich mit menschlichen Bewertern validiert worden sind.

Fazitbox

• Bei richtiger Anwendung kann ein KI-basiertes Screening-Tool dazu beitragen, unnötige medizinische Untersuchungen zu reduzieren und die Verfügbarkeit medizinischer Versorgung auf die Patienten zu konzentrieren, die sie am dringendsten benötigen.

• Die ordnungsgemäße Implementierung von KI-basiertem DR-Screening stellt vielerorts noch eine technische und logistische Herausforderung dar, hat jedoch enormes Potenzial, die Sehfähigkeit vieler Menschen zu erhalten. Die mangelnde Transparenz hinsichtlich der Entscheidungsfindung von KI-Modellen stellt jedoch eine essenzielle Limitation für deren praktische Nutzung dar.

• KI-Modelle können vielseitig eingesetzt werden und bieten insbesondere in Regionen mit einem Mangel an Fachärzten wertvolle Unterstützung zur Etablierung möglicher Triage-Systeme, um die Patientenversorgung zu strukturieren. Die Augenheilkunde als Disziplin mit einem starken Fokus auf optische Verfahren und stetiger Weiterentwicklung nicht invasiver Bildgebungstechniken bietet zahlreiche potenzielle Anwendungsbereiche für automatische oder semiautomatische KI-Modelle.

Conclusion Box

• When used properly, an AI-based screening tool can help reduce unneeded medical examinations and focus health care on the patients who need it most urgently.

• Implementing AI-based DR screening properly is still a technical and logistical challenge in many places, but it has huge potential to sustain the eyesight of many people. However, the lack of transparency in AI modelsʼ decision-making process represents a significant limitation for their practical use.

• AI models can be used in a variety of ways and offer valuable support for establishing possible triage systems to structure patient care, especially in regions with a shortage of specialists. Ophthalmology, a field with a strong focus on optic procedures and continuous development of non-invasive imaging techniques, offers numerous potential applications for automated or semi-automated AI modelling.

Schlüsselwörter

Informationstechnologie - Geschichte der Medizin - Retina - KI - diabetische Retinopathie

Publication History

Received: 06 October 2024

Accepted: 12 February 2025

Article published online:
02 June 2025

• References/Literatur

• 1 International Diabetes Federation. DF Diabetes Atlas 2021. 10th ed. https://diabetesatlas.org/atlas/tenth-edition/ Stand: 25.03.2025
• 2 Fong DS, Aiello L, Gardner TW. et al. Retinopathy in Diabetes. Diabetes Care 2004; 27: s84-s87
• 3 Teo ZL, Tham YC, Yu M. et al. Global Prevalence of Diabetic Retinopathy and Projection of Burden through 2045: Systematic Review and Meta-analysis. Ophthalmology 2021; 128: 1580-1591
• 4 Bundesärztekammer, Kassenärztliche Bundesvereinigung, Arbeitsgemeinschaft der Wissenschaftlichen Medizinischen Fachgesellschaften. Nationale VersorgungsLeitlinie (NVL) Typ-2-Diabetes. 15.05.2023. Langfassung – Version 3.0. AWMF-Register-Nr. nvl-001. https://register.awmf.org/assets/guidelines/nvl-001l_S3_Typ-2-Diabetes_2023-05.pdf Stand: 25.03.2025
• 5 Ferris FL3rd. How Effective Are Treatments for Diabetic Retinopathy?. JAMA 1993; 269: 1290-1291
• 6 Kreft D, McGuinness MB, Doblhammer G. et al. Diabetic retinopathy screening in incident diabetes mellitus type 2 in Germany between 2004 and 2013 – A prospective cohort study based on health claims data. PLoS One 2018; 13: e0195426
• 7 Spital G, Faatz H. Diabetic Retinopathy – a Common Disease. Klin Monbl Augenheilkd 2023; 240: 1060-1070
• 8 Whitestone N, Nkurikiye J, Patnaik JL. et al. Feasibility and acceptance of artificial intelligence-based diabetic retinopathy screening in Rwanda. Br J Ophthalmol 2024; 108: 840-845
• 9 Abràmoff MD, Lavin PT, Birch M. et al. Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices. NPJ Digit Med 2018; 1: 39
• 10 Ting DSW, Cheung CY, Lim G. et al. Development and Validation of a Deep Learning System for Diabetic Retinopathy and Related Eye Diseases Using Retinal Images From Multiethnic Populations With Diabetes. JAMA 2017; 318: 2211-2223
• 11 Gargeya R, Leng T. Automated Identification of Diabetic Retinopathy Using Deep Learning. Ophthalmology 2017; 124: 962-969
• 12 Gulshan V, Peng L, Coram M. et al. Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs. JAMA 2016; 316: 2402-2410
• 13 Rajesh AE, Davidson OQ, Lee CS. et al. Artificial Intelligence and Diabetic Retinopathy: AI Framework, Prospective Studies, Head-to-head Validation, and Cost-effectiveness. Diabetes Care 2023; 46: 1728-1739
• 14 Li F, Wang Y, Xu T. et al. Deep learning-based automated detection for diabetic retinopathy and diabetic macular oedema in retinal fundus photographs. Eye (Lond) 2022; 36: 1433-1441
• 15 Joseph S, Selvaraj J, Mani I. et al. Diagnostic Accuracy of Artificial Intelligence-Based Automated Diabetic Retinopathy Screening in Real-World Settings: A Systematic Review and Meta-Analysis. Am J Ophthalmol 2024; 263: 214-230
• 16 Nørgaard MF, Grauslund J. Automated Screening for Diabetic Retinopathy – A Systematic Review. Ophthalmic Res 2018; 60: 9-17
• 17 Uy H, Fielding C, Hohlfeld A. et al. Diagnostic test accuracy of artificial intelligence in screening for referable diabetic retinopathy in real-world settings: A systematic review and meta-analysis. PLOS Global Public Health 2023; 3: e0002160
• 18 Roser P, Grohmann C, Aberle J. et al. Evaluation der Implementierung eines zugelassenen Künstliche-Intelligenz-Systems zur Erkennung der diabetischen Retinopathie. Diabetologie und Stoffwechsel 2021; 16: 402-408
• 19 van der Heijden AA, Abramoff MD, Verbraak F. et al. Validation of automated screening for referable diabetic retinopathy with the IDx-DR device in the Hoorn Diabetes Care System. Acta Ophthalmol 2018; 96: 63-68
• 20 Paul S, Tayar A, Morawiec-Kisiel E. et al. [Use of artificial intelligence in screening for diabetic retinopathy at a tertiary diabetes center]. Ophthalmologie 2022; 119: 705-713
• 21 Bellemo V, Lim ZW, Lim G. et al. Artificial intelligence using deep learning to screen for referable and vision-threatening diabetic retinopathy in Africa: a clinical validation study. Lancet Digit Health 2019; 1: e35-e44
• 22 Xie Y, Nguyen QD, Hamzah H. et al. Artificial intelligence for teleophthalmology-based diabetic retinopathy screening in a national programme: an economic analysis modelling study. Lancet Digit Health 2020; 2: e240-e249
• 23 Skevas C, de Olaguer NP, Lleó A. et al. Implementing and evaluating a fully functional AI-enabled model for chronic eye disease screening in a real clinical environment. BMC Ophthalmol 2024; 24: 51
• 24 Browning DJ, Stewart MW, Lee C. Diabetic macular edema: Evidence-based management. Indian J Ophthalmol 2018; 66: 1736-1750
• 25 Mackenzie S, Schmermer C, Charnley A. et al. SDOCT Imaging to Identify Macular Pathology in Patients Diagnosed with Diabetic Maculopathy by a Digital Photographic Retinal Screening Programme. PLoS One 2011; 6: e14811
• 26 Grauslund J, Andersen N, Andresen J. et al. Evidence-based Danish guidelines for screening of diabetic retinopathy. Acta Ophthalmol 2018; 96: 763-769
• 27 Kermany DS, Goldbaum M, Cai W. et al. Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning. Cell 2018; 172: 1122-1131.e9
• 28 Ye X, Gao K, He S. et al. Artificial Intelligence-Based Quantification of Central Macular Fluid Volume and VA Prediction for Diabetic Macular Edema Using OCT Images. Ophthalmol Ther 2023; 12: 2441-2452
• 29 Zhang Y, Xu F, Lin Z. et al. Prediction of Visual Acuity after anti-VEGF Therapy in Diabetic Macular Edema by Machine Learning. J Diabetes Res 2022; 2022: 5779210
• 30 Quek TC, Takahashi K, Kang HG. et al. Predictive, preventive, and personalized management of retinal fluid via computer-aided detection app for optical coherence tomography scans. EPMA J 2022; 13: 547-560
• 31 Tan TE, Wong TY, Ting DSW. Artificial Intelligence for Prediction of Anti-VEGF Treatment Burden in Retinal Diseases: Towards Precision Medicine. Ophthalmol Retina 2021; 5: 601-603
• 32 Bajka A, Bacci T, Wiest MRJ. et al. Feasibility and Clinical Utility of Wide-Field Optical Coherence Tomography Angiography Compared to Ultrawide-Field Fluorescein Angiography in Patients with Diabetic Retinopathy. Klin Monbl Augenheilkd 2023; 240: 490-495
• 33 Wen J, Liu D, Wu Q. et al. Retinal image-based artificial intelligence in detecting and predicting kidney diseases: Current advances and future perspectives. VIEW 2023; 4: 20220070
• 34 Zhang K, Liu X, Xu J. et al. Deep-learning models for the detection and incidence prediction of chronic kidney disease and type 2 diabetes from retinal fundus images. Nat Biomed Eng 2021; 5: 533-545
• 35 Schubert M. The Correlation between Diabetes Mellitus and Neurodegenerative Diseases. Klin Monbl Augenheilkd 2023; 240: 130-135
• 36 Kopf D, Frölich L. Risk of incident Alzheimerʼs disease in diabetic patients: a systematic review of prospective trials. J Alzheimers Dis 2009; 16: 677-685
• 37 Chatterjee S, Peters SA, Woodward M. et al. Type 2 Diabetes as a Risk Factor for Dementia in Women Compared With Men: A Pooled Analysis of 2.3 Million People Comprising More Than 100,000 Cases of Dementia. Diabetes Care 2016; 39: 300-307
• 38 Zhang J, Chen C, Hua S. et al. An updated meta-analysis of cohort studies: Diabetes and risk of Alzheimerʼs disease. Diabetes Res Clin Pract 2017; 124: 41-47
• 39 Huang Y, Wang S, Cai C. et al. Retinal vascular density as a potential biomarker of diabetic cerebral small vessel disease. Diabetes Obes Metab 2024; 26: 1789-1798
• 40 Majimbi M, McLenachan S, Nesbit M. et al. In vivo retinal imaging is associated with cognitive decline, blood-brain barrier disruption and neuroinflammation in type 2 diabetic mice. Frontiers in Endocrinology 2023; 14
• 41 Ciudin A, Simó-Servat O, Hernández C. et al. Retinal Microperimetry: A New Tool for Identifying Patients With Type 2 Diabetes at Risk for Developing Alzheimer Disease. Diabetes 2017; 66: 3098-3104
• 42 Liang M, Chen X, Xue F. et al. Diffusion-weighted imaging of injuries to the visual centers of the brain in patients with type 2 diabetes and retinopathy. Exp Ther Med 2017; 14: 1153-1156
• 43 Grzybowski A, Singhanetr P, Nanegrungsunk O. et al. Artificial Intelligence for Diabetic Retinopathy Screening Using Color Retinal Photographs: From Development to Deployment. Ophthalmol Ther 2023; 12: 1419-1437
• 44 Tufail A, Rudisill C, Egan C. et al. Automated Diabetic Retinopathy Image Assessment Software: Diagnostic Accuracy and Cost-Effectiveness Compared with Human Graders. Ophthalmology 2017; 124: 343-351
• 45 [Anonym] Art. 22 DSGVO: Automatisierte Entscheidungen im Einzelfall einschließlich Profiling. https://dsgvo-gesetz.de/art-22-dsgvo/ Stand: 25.03.2025
• 46 McMaster University, Evidencde Prime Inc.. GRADEpro GDT: GRADEpro Guideline Development Tool 2022. https://gdt.gradepro.org/app/ Stand: 25.03.2025
• 47 Poschkamp B, Stahl A. Application of deep learning algorithms for diabetic retinopathy screening. Ann Transl Med 2022; 10: 1298
• 48 Lundeen EA, Andes LJ, Rein DB. et al. Trends in Prevalence and Treatment of Diabetic Macular Edema and Vision-Threatening Diabetic Retinopathy Among Medicare Part B Fee-for-Service Beneficiaries. JAMA Ophthalmol 2022; 140: 345-353
• 49 Wintergerst MWM, Bejan V, Hartmann V. et al. Telemedical Diabetic Retinopathy Screening in a Primary Care Setting: Quality of Retinal Photographs and Accuracy of Automated Image Analysis. Ophthalmic Epidemiol 2022; 29: 286-295
• 50 Zech H, Hünefeld IC. Einsatz von KI in der Medizin: Haftung und Versicherung. Medizinrecht 2023; 41: 1-8
• 51 Wehkamp K, Krawczak M, Schreiber S. The Quality and Utility of Artificial Intelligence in Patient Care. Dtsch Arztebl Int 2023; 120: 463-469
• 52 Cabitza F, Campagner A, Balsano C. Bridging the “last mile” gap between AI implementation and operation: “data awareness” that matters. Ann Transl Med 2020; 8: 501
• 53 Arcadu F, Benmansour F, Maunz A. et al. Deep learning algorithm predicts diabetic retinopathy progression in individual patients. NPJ Digit Med 2019; 2: 92