EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Vision transformer with masked autoencoders for referable diabetic retinopathy classification based on large-size retina image

Yaoming Yang iD, Zhili Cai, Shuxia Qiu, Peng Xu iD

PLOS ONE · 06.03.2024 · 19 (3) · e0299265

DOI: 10.1371/journal.pone.0299265 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS ONE, online erschienen am 06.03.2024. Verfasst von Yaoming Yang und 3 weiteren Autor:innen. Beteiligte Einrichtung: College of Science, China Jiliang University, Hangzhou, Zhejiang, China. Hauptthemen laut MEDLINE: Diabetic Retinopathy, Diabetes Mellitus. Die Arbeit wurde bisher 68-mal zitiert (Crossref) und nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Computer-aided diagnosis systems based on deep learning algorithms have shown potential applications in rapid diagnosis of diabetic retinopathy (DR). Due to the superior performance of Transformer over convolutional neural networks (CNN) on natural images, we attempted to develop a new model to classify referable DR based on a limited number of large-size retinal images by using Transformer. Vision Transformer (ViT) with Masked Autoencoders (MAE) was applied in this study to improve the classification performance of referable DR. We collected over 100,000 publicly fundus retinal images larger than 224×224, and then pre-trained ViT on these retinal images using MAE. The pre-trained ViT was applied to classify referable DR, the performance was also compared with that of ViT pre-trained using ImageNet. The improvement in model classification performance by pre-training with over 100,000 retinal images using MAE is superior to that pre-trained with ImageNet. The accuracy, area under curve (AUC), highest sensitivity and highest specificity of the present model are 93.42%, 0.9853, 0.973 and 0.9539, respectively. This study shows that MAE can provide more flexibility to the input image and substantially reduce the number of images required. Meanwhile, the pretraining dataset scale in this study is much smaller than ImageNet, and the pre-trained weights from ImageNet are not required also.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Diabetic RetinopathyDiabetes MellitusAnimalsAbomasumAlgorithmsArea Under CurveFundus Oculi

Autor:innen und Institutionen

  • College of Science, China Jiliang University, Hangzhou, Zhejiang, China.
  • Key Laboratory of Intelligent Manufacturing Quality Big Data Tracing and Analysis of Zhejiang Province, Hangzhou, Zhejiang, China.

ORCID-Profile: Yaoming Yang, Peng Xu

Förderung

  • Natural Science Foundation of China

Zitieren

APA

Yang, Y., Cai, Z., Qiu, S., & Xu, P. (2024). Vision transformer with masked autoencoders for referable diabetic retinopathy classification based on large-size retina image. PLOS ONE, 19(3), e0299265. https://doi.org/10.1371/journal.pone.0299265

Vancouver

Yang Y, Cai Z, Qiu S, Xu P. Vision transformer with masked autoencoders for referable diabetic retinopathy classification based on large-size retina image. PLOS ONE. 2024;19(3):e0299265. doi:10.1371/journal.pone.0299265

BibTeX
@article{Yang2024,
  author = {Yang, Yaoming and Cai, Zhili and Qiu, Shuxia and Xu, Peng},
  title = {Vision transformer with masked autoencoders for referable diabetic retinopathy classification based on large-size retina image},
  journal = {PLOS ONE},
  year = {2024},
  volume = {19},
  number = {3},
  pages = {e0299265},
  doi = {10.1371/journal.pone.0299265}
}
Kostenlos starten →