The AROMA Dataset for Automatic Detection of Artifact Type and Severity in Retinal Optical Coherence Tomography Angiography
Ophthalmic Research · 05.05.2026 · 69 (1) · 105-119
DOI: 10.1159/000551126 ↗Kurz gefasst
Fachartikel in der Zeitschrift Ophthalmic Research, online erschienen am 05.05.2026. Verfasst von Lucas Szwarcberg und 7 weiteren Autor:innen. Beteiligte Einrichtung: Department of Ophthalmology, Dijon University Hospital, Université Bourgogne Europe, Dijon, France. Hauptthemen laut MEDLINE: Tomography, Optical Coherence, Artifacts, Retinal Vessels, Fluorescein Angiography. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Introduction
In recent years, retinal vascular imaging has attracted growing interest and is experiencing rapid technological advancements in imaging modalities such as swept-source optical coherence tomography angiography (SS OCT-A). OCT-A enables precise, noninvasive, quantitative measurements of retinal vascularization. However, it is also prone to artifacts that are challenging to detect and can significantly limit diagnostic accuracy and biomarker reliability.
Methods
We present a dataset of en face retinal SS OCT-A images, labeled by artifact type and severity. Each patient’s OCT-A scan consists of 14 images that are graded by multiple experts and assigned corresponding artifact labels with an overall quality grade. We also report results for two deep-learning binary classification models trained on the dataset: one based on image compression via principal component analysis (PCA) and the other using a six-channel tensor.
Results
The dataset comprised 281 OCT-A scans from 115 anonymized patients. The PCA-based model achieved a precision of 87% for image quality classification, while the six-channel model achieved 97%.
Conclusion
Using a curated dataset of en face SS OCT-A images with detailed artifact annotations, we trained a deep-learning model capable of high-precision image quality classification. These results demonstrate the feasibility of using task-specific artificial intelligence for quality assessment based on artifact detection in retinal imaging.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Ophthalmology, Dijon University Hospital, Université Bourgogne Europe, Dijon, France.
- ICMUB UMR 6302, CNRS, Université Bourgogne Europe, Dijon, France.
- Department of Anaesthesiology, APHM North University Hospital, Marseille, France.
- INRAE, Dijon University Hospital, Université Bourgogne Europe, Dijon, France.
- Department of Ophthalmology, Dijon University Hospital, Université Bourgogne Europe, Dijon, France, louis.arnould@chu-dijon.fr.
- PEC2, Dijon University Hospital, Université Bourgogne Europe, Dijon, France, louis.arnould@chu-dijon.fr.
Zitieren
APA
Szwarcberg, L., Anwer, A., Gozlan, A., Drouard, P., Gabrielle, P. H., Creuzot-Garcher, C., Mériaudeau, F., & Arnould, L. (2026). The AROMA Dataset for Automatic Detection of Artifact Type and Severity in Retinal Optical Coherence Tomography Angiography. Ophthalmic Research, 69(1), 105-119. https://doi.org/10.1159/000551126
Vancouver
Szwarcberg L, Anwer A, Gozlan A, Drouard P, Gabrielle PH, Creuzot-Garcher C, et al. The AROMA Dataset for Automatic Detection of Artifact Type and Severity in Retinal Optical Coherence Tomography Angiography. Ophthalmic Research. 2026;69(1):105-119. doi:10.1159/000551126
BibTeX
@article{Szwarcberg2026,
author = {Szwarcberg, Lucas and Anwer, Atif and Gozlan, Alexandre and Drouard, Paul and Gabrielle, Pierre-Henry and Creuzot-Garcher, Catherine and Mériaudeau, Fabrice and Arnould, Louis},
title = {The AROMA Dataset for Automatic Detection of Artifact Type and Severity in Retinal Optical Coherence Tomography Angiography},
journal = {Ophthalmic Research},
year = {2026},
volume = {69},
number = {1},
pages = {105--119},
doi = {10.1159/000551126}
}
EUVIMED