Abstract
Image quality remains a key problem for both traditional and deep learning (DL)-based approaches to retinal image analysis and identifying poor quality images can be time consuming and subjective. Thus, automated methods for retinal image quality scoring (RIQS) are needed. The current state-of-the-art is MCFNet, composed of three Densenet121 backbones each operating in a different colour space. MCFNet, and the EyeQ dataset released by the same authors, was a huge step forward for RIQS. We present QuickQual, a simple approach to RIQS, consisting of a single “off-the-shelf” ImageNet-pretrained Densenet121 backbone plus a Support Vector Machine (SVM). QuickQual performs very well, setting a new state-of-the-art for EyeQ (Accuracy: 88.50% vs 88.00% for MCFNet; AUC: 0.9687 vs 0.9588). This suggests that RIQS can be solved with generic “perceptual” features learned on natural images, as opposed to requiring DL models trained on large amounts of fundus images. Additionally, we propose a Fixed Prior linearisation scheme, that converts EyeQ from a 3-class classification to a continuous logistic regression task. For this task, we present a second model, QuickQual MEga Minified Estimator (QuickQual-MEME), that consists of only 10 parameters on top of an off-the-shelf Densenet121 and can distinguish between gradable and ungradable images with an accuracy of 89.18% (AUC: 0.9537). Code and model are available on GitHub. QuickQual is so lightweight, that the entire inference code (and even the parameters for QuickQual-MEME) is already contained in this paper.
A. Storkey and M. O. Bernabeu—Equal supervision.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Note that for MCFNet, the original accuracy scores provided were not entirely accurate due to a bug in the evaluation code. See the note here on the Github for MCFNet: https://github.com/HzFu/EyeQ#-reference “Note: The corrected accuracy score of MCF-Net is 0.8800.” We thank the authors of MCFNet for their exceptional transparency in sharing not just code, model weights and data, but also their model’s test set predictions.
References
Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., Fei-Fei, L.: ImageNet: a large-scale hierarchical image database. In: 2009 IEEE Conference on Computer Vision and Pattern Recognition, pp. 248–255. IEEE (2009)
Engelmann, J., Storkey, A., Bernabeu, M.O.: Deep learning (dl) identifies age as key axis of perceptual variation in fundus images-without training on fundus images. Investigat. Ophthalmol. Vis. Sci. 64(9), PB004 (2023)
Engelmann, J., Storkey, A., LLinares, M.B.: Exclusion of poor quality fundus images biases health research linking retinal traits and systemic health. Investigat. Ophthalmol. Vis. Sci. 64(8), 2922 (2023)
Engelmann, J., Villaplana-Velasco, A., Storkey, A., Bernabeu, M.O.: Robust and efficient computation of retinal fractal dimension through deep approximation. In: International Workshop on Ophthalmic Medical Image Analysis, pp. 84–93. Springer, Cham (2022). https://doi.org/10.1007/978-3-031-16525-2_9
Fu, H., et al.: Evaluation of retinal image quality assessment networks in different color-spaces. In: Shen, D., et al. (eds.) MICCAI 2019. LNCS, vol. 11764, pp. 48–56. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-32239-7_6
Huang, G., Liu, Z., Van Der Maaten, L., Weinberger, K.Q.: Densely connected convolutional networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4700–4708 (2017)
Laurik-Feuerstein, K.L., Sapahia, R., Cabrera DeBuc, D., Somfai, G.M.: The assessment of fundus image quality labeling reliability among graders with different backgrounds. PLoS One 17(7), e0271156 (2022)
MacGillivray, T.J., et al.: Suitability of UK biobank retinal images for automatic analysis of morphometric properties of the vasculature. PLoS One 10(5), e0127914 (2015)
MacGillivray, T., Trucco, E., Cameron, J., Dhillon, B., Houston, J., Van Beek, E.: Retinal imaging as a source of biomarkers for diagnosis, characterization and prognosis of chronic illness or long-term conditions. Br. J. Radiol. 87(1040), 20130832 (2014)
Shen, Z., Fu, H., Shen, J., Shao, L.: Modeling and enhancing low-quality retinal fundus images. IEEE Trans. Med. Imag. 40(3), 996–1006 (2020)
Velasco, A.V., et al.: Decreased retinal vascular complexity is an early biomarker of mi supported by a shared genetic control. medRxiv (2021)
Wagner, S.K., et al.: Insights into systemic disease through retinal imaging-based oculomics. Transl. Vis. Sci. Technol. 9(2), 6 (2020)
Wightman, R.: PyTorch Image Models (2019). https://github.com/rwightman/pytorch-image-models. https://doi.org/10.5281/zenodo.4414861
Zekavat, S.M., et al.: Deep learning of the retina enables phenome-and genome-wide analyses of the microvasculature. Circulation 145(2), 134–150 (2022)
Zhou, Y., et al.: Automorph: automated retinal vascular morphology quantification via a deep learning pipeline. Transl. Vis. Sci. Technol. 11(7), 12 (2022)
Acknowledgements
We thank our friends and colleagues for their help and support. J.E. and this work was supported by the United Kingdom Research and Innovation (grant EP/S02431X/1), UKRI Centre for Doctoral Training in Biomedical AI at the University of Edinburgh, School of Informatics. For the purpose of open access, the author has applied a creative commons attribution (CC BY) licence to any author accepted manuscript version arising.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Engelmann, J., Storkey, A., Bernabeu, M.O. (2023). QuickQual: Lightweight, Convenient Retinal Image Quality Scoring with Off-the-Shelf Pretrained Models. In: Antony, B., Chen, H., Fang, H., Fu, H., Lee, C.S., Zheng, Y. (eds) Ophthalmic Medical Image Analysis. OMIA 2023. Lecture Notes in Computer Science, vol 14096. Springer, Cham. https://doi.org/10.1007/978-3-031-44013-7_4
Download citation
DOI: https://doi.org/10.1007/978-3-031-44013-7_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-44012-0
Online ISBN: 978-3-031-44013-7
eBook Packages: Computer ScienceComputer Science (R0)
