EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

DeepAlienorNet: A deep learning model to extract clinical features from colour fundus photography in age‐related macular degeneration

Alexis Mathieu iD, Soufiane Ajana, Jean-François Korobelnik iD, Mélanie Le Goff, Brigitte Gontier, Marie-Bénédicte Rougier, Cécile Delcourt iD, Marie-Noëlle Delyfer iD

Acta Ophthalmologica · 2024

DOI: 10.1111/aos.16660 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Acta Ophthalmologica, 2024. Verfasst von Alexis Mathieu und 7 weiteren Autor:innen. Beteiligte Einrichtung: Inserm, Bordeaux Population Health Research Center, UMR 1219, University of Bordeaux, Bordeaux, France. Hauptthemen laut MEDLINE: Deep Learning, Photography, Fundus Oculi. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Ministère de la Santé, Agence Nationale de la Recherche, CFSR Recherche (Club Francophone des Spécialistes de la Rétine).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Objective

This study aimed to develop a deep learning (DL) model, named ‘DeepAlienorNet’, to automatically extract clinical signs of age‐related macular degeneration (AMD) from colour fundus photography (CFP). Methods and Analysis The ALIENOR Study is a cohort of French individuals 77 years of age or older. A multi‐label DL model was developed to grade the presence of 7 clinical signs: large soft drusen (>125 μm), intermediate soft (63–125 μm), large area of soft drusen (total area >500 μm), presence of central soft drusen (large or intermediate), hyperpigmentation, hypopigmentation, and advanced AMD (defined as neovascular or atrophic AMD). Prediction performances were evaluated using cross‐validation and the expert human interpretation of the clinical signs as the ground truth.

Results

A total of 1178 images were included in the study. Averaging the 7 clinical signs' detection performances, DeepAlienorNet achieved an overall sensitivity, specificity, and AUROC of 0.77, 0.83, and 0.87, respectively. The model demonstrated particularly strong performance in predicting advanced AMD and large areas of soft drusen. It can also generate heatmaps, highlighting the relevant image areas for interpretation.

Conclusion

DeepAlienorNet demonstrates promising performance in automatically identifying clinical signs of AMD from CFP, offering several notable advantages. Its high interpretability reduces the black box effect, addressing ethical concerns. Additionally, the model can be easily integrated to automate well‐established and validated AMD progression scores, and the user‐friendly interface further enhances its usability. The main value of DeepAlienorNet lies in its ability to assist in precise severity scoring for further adapted AMD management, all while preserving interpretability.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningPhotographyFundus OculiHumansAgedMaleFemaleMacular DegenerationRetinal DrusenAged, 80 and overROC Curve

Stichwörter der Autor:innen

artificial intelligencecomputer‐assisteddeep learningfundus photographyimage interpretationmacular degeneration

Autor:innen und Institutionen

  • Inserm, Bordeaux Population Health Research Center, UMR 1219, University of Bordeaux, Bordeaux, France.
  • Service d'Ophtalmologie, Centre Hospitalier Universitaire de Bordeaux, Bordeaux, France.
  • FRCRnet/FCRIN Network, Bordeaux, France.

ORCID-Profile: Alexis Mathieu, Jean-François Korobelnik, Cécile Delcourt, Marie-Noëlle Delyfer

Förderung

  • Ministère de la Santé
  • Agence Nationale de la Recherche
  • CFSR Recherche (Club Francophone des Spécialistes de la Rétine)
  • University of Bordeaux
  • Fondation Voir et Entendre
  • Théa Pharma
  • CNSA (Caisse Nationale pour la Solidarité et l'Autonomie)
  • Université de Bordeaux

Zitieren

APA

Mathieu, A., Ajana, S., Korobelnik, J. F., Le Goff, M., Gontier, B., Rougier, M. B., Delcourt, C., & Delyfer, M. N. (2024). DeepAlienorNet: A deep learning model to extract clinical features from colour fundus photography in age‐related macular degeneration. Acta Ophthalmologica. https://doi.org/10.1111/aos.16660

Vancouver

Mathieu A, Ajana S, Korobelnik JF, Le Goff M, Gontier B, Rougier MB, et al. DeepAlienorNet: A deep learning model to extract clinical features from colour fundus photography in age‐related macular degeneration. Acta Ophthalmologica. 2024. doi:10.1111/aos.16660

BibTeX
@article{Mathieu2024,
  author = {Mathieu, Alexis and Ajana, Soufiane and Korobelnik, Jean-François and Le Goff, Mélanie and Gontier, Brigitte and Rougier, Marie-Bénédicte and Delcourt, Cécile and Delyfer, Marie-Noëlle},
  title = {DeepAlienorNet: A deep learning model to extract clinical features from colour fundus photography in age‐related macular degeneration},
  journal = {Acta Ophthalmologica},
  year = {2024},
  doi = {10.1111/aos.16660}
}
Kostenlos starten →