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

Detecting Glaucoma from Fundus Photographs Using Deep Learning without Convolutions

Rui Fan, Kamran Alipour, Christopher Bowd, Mark Christopher, Nicole Brye, James A Proudfoot, Michael H Goldbaum, Akram Belghith, Christopher A Girkin, Massimo A Fazio, Jeffrey M Liebmann, Robert N Weinreb und 3 weitere

Ophthalmology Science · 2023-03 · 3 (1) · 100233

DOI: 10.1016/j.xops.2022.100233 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Ophthalmology Science, 2023-03. Verfasst von Rui Fan und 14 weiteren Autor:innen. Beteiligte Einrichtung: Hamilton Glaucoma Center, Viterbi Family Department of Ophthalmology and Shiley Eye Institute, University of California San Diego, La Jolla, Californ…. Die Arbeit wurde bisher 135-mal zitiert (Crossref) und nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: NEI NIH HHS, National Eye Institute, Research to Prevent Blindness.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

To compare the diagnostic accuracy and explainability of a Vision Transformer deep learning technique, Data-efficient image Transformer (DeiT), and ResNet-50, trained on fundus photographs from the Ocular Hypertension Treatment Study (OHTS) to detect primary open-angle glaucoma (POAG) and identify the salient areas of the photographs most important for each model's decision-making process.

Design

Evaluation of a diagnostic technology. Subjects Participants and Controls: Overall 66 715 photographs from 1636 OHTS participants and an additional 5 external datasets of 16 137 photographs of healthy and glaucoma eyes.

Methods

Data-efficient image Transformer models were trained to detect 5 ground-truth OHTS POAG classifications: OHTS end point committee POAG determinations because of disc changes (model 1), visual field (VF) changes (model 2), or either disc or VF changes (model 3) and Reading Center determinations based on disc (model 4) and VFs (model 5). The best-performing DeiT models were compared with ResNet-50 models on OHTS and 5 external datasets. Main Outcome Measures: Diagnostic performance was compared using areas under the receiver operating characteristic curve (AUROC) and sensitivities at fixed specificities. The explainability of the DeiT and ResNet-50 models was compared by evaluating the attention maps derived directly from DeiT to 3 gradient-weighted class activation map strategies.

Results

Compared with our best-performing ResNet-50 models, the DeiT models demonstrated similar performance on the OHTS test sets for all 5 ground-truth POAG labels; AUROC ranged from 0.82 (model 5) to 0.91 (model 1). Data-efficient image Transformer AUROC was consistently higher than ResNet-50 on the 5 external datasets. For example, AUROC for the main OHTS end point (model 3) was between 0.08 and 0.20 higher in the DeiT than ResNet-50 models. The saliency maps from the DeiT highlight localized areas of the neuroretinal rim, suggesting important rim features for classification. The same maps in the ResNet-50 models show a more diffuse, generalized distribution around the optic disc.

Conclusions

Vision Transformers have the potential to improve generalizability and explainability in deep learning models, detecting eye disease and possibly other medical conditions that rely on imaging for clinical diagnosis and management.

Schlagwörter

Stichwörter der Autor:innen

AI, artificial intelligenceAUROC, areas under the receiver operating characteristic curveCI, confidence intervalCNN, convolutional neural networkDL, deep learningDeep learningDeiT, Data-efficient image TransformerFundus photographsGlaucoma detectionLAG, Large-Scale Attention-Based GlaucomaOHTS, Ocular Hypertension Treatment StudyPOAG, primary open-angle glaucomaSoTA, state-of-the-artVF, visual fieldViT, Vision Transformer

Autor:innen und Institutionen

  • Hamilton Glaucoma Center, Viterbi Family Department of Ophthalmology and Shiley Eye Institute, University of California San Diego, La Jolla, California.
  • Department of Computer Science and Engineering, University of California San Diego, La Jolla, California.
  • Department of Control Science and Engineering, Tongji University, Shanghai 201804, China.
  • Department of Ophthalmology, School of Medicine, The University of Alabama at Birmingham, Birmingham, Alabama.
  • Department of Biomedical Engineering, School of Engineering, The University of Alabama at Birmingham, Birmingham, Alabama.
  • Bernard and Shirlee Brown Glaucoma Research Laboratory, Edward S. Harkness Eye Institute, Columbia University Medical Center, New York, New York.

Förderung

  • NEI NIH HHS
  • National Eye Institute
  • Research to Prevent Blindness

Zitieren

APA

Fan, R., Alipour, K., Bowd, C., Christopher, M., Brye, N., Proudfoot, J. A., Goldbaum, M. H., Belghith, A., Girkin, C. A., Fazio, M. A., Liebmann, J. M., Weinreb, R. N., Pazzani, M., Kriegman, D., & Zangwill, L. M. (2023). Detecting Glaucoma from Fundus Photographs Using Deep Learning without Convolutions. Ophthalmology Science, 3(1), 100233. https://doi.org/10.1016/j.xops.2022.100233

Vancouver

Fan R, Alipour K, Bowd C, Christopher M, Brye N, Proudfoot JA, et al. Detecting Glaucoma from Fundus Photographs Using Deep Learning without Convolutions. Ophthalmology Science. 2023;3(1):100233. doi:10.1016/j.xops.2022.100233

BibTeX
@article{Fan2023,
  author = {Fan, Rui and Alipour, Kamran and Bowd, Christopher and Christopher, Mark and Brye, Nicole and Proudfoot, James A and Goldbaum, Michael H and Belghith, Akram and Girkin, Christopher A and Fazio, Massimo A and Liebmann, Jeffrey M and Weinreb, Robert N and Pazzani, Michael and Kriegman, David and Zangwill, Linda M},
  title = {Detecting Glaucoma from Fundus Photographs Using Deep Learning without Convolutions},
  journal = {Ophthalmology Science},
  year = {2023},
  volume = {3},
  number = {1},
  pages = {100233},
  doi = {10.1016/j.xops.2022.100233}
}
Kostenlos starten →