Image preprocessing‐based ensemble deep learning classification of diabetic retinopathy
IET Image Processing · 2024-02 · 18 (3) · 807-828
DOI: 10.1049/ipr2.12987 ↗Kurz gefasst
Fachartikel in der Zeitschrift IET Image Processing, online erschienen am 15.11.2023. Verfasst von Peter Macsik und 4 weiteren Autor:innen. Die Arbeit wurde bisher 38-mal zitiert (Crossref) und nennt 77 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: European Regional Development Fund.
Abstract (Originalsprache)
Diabetic retinopathy (DR) can cause irreversible eye damage, even blindness. The prognosis improves with early diagnosis. According to the International Classification of Diabetic Retinopathy Severity Scale (ICDRSS), DR has five stages. Modern, cost‐effective techniques for automatic DR screening and staging of fundus images are based on deep learning (DL). To obtain higher classification accuracy, the combination of several diverse individual DL models into one ensemble could be used. A new approach to model diversity in an ensemble is proposed by manipulating the training input data involving original and four variants of preprocessed image datasets. There are publicly available datasets with labels for all five stages, but some contain poor‐quality images. In contrast, this algorithm was trained on images from a six‐class DDR dataset, including the class of poor‐quality ungradable images, to enhance the classification performance. The solution was evaluated on the APTOS dataset, containing only ICDRSS classes. Classification results of the ensemble model were presented on two different ensemble convolutional neural network (CNN) models, based on Xception and EfficientNetB4 architectures using two fusion approaches. Our proposed ensemble models outperformed all other single deep learning architectures regarding overall accuracy and Cohen's Kappa, with the best results using the EfficientNetB4 architecture.
Förderung
- European Regional Development Fund
Zitieren
APA
Macsik, P., Pavlovicova, J., Kajan, S., Goga, J., & Kurilova, V. (2024). Image preprocessing‐based ensemble deep learning classification of diabetic retinopathy. IET Image Processing, 18(3), 807-828. https://doi.org/10.1049/ipr2.12987
Vancouver
Macsik P, Pavlovicova J, Kajan S, Goga J, Kurilova V. Image preprocessing‐based ensemble deep learning classification of diabetic retinopathy. IET Image Processing. 2024;18(3):807-828. doi:10.1049/ipr2.12987
BibTeX
@article{Macsik2024,
author = {Macsik, Peter and Pavlovicova, Jarmila and Kajan, Slavomir and Goga, Jozef and Kurilova, Veronika},
title = {Image preprocessing‐based ensemble deep learning classification of diabetic retinopathy},
journal = {IET Image Processing},
year = {2024},
volume = {18},
number = {3},
pages = {807--828},
doi = {10.1049/ipr2.12987}
}
EUVIMED