Abstract
Evaluating optic nerve cross-sectional images requires significant time and effort by experts to perform. Autonomous systems using deep learning can help to reduce this workload by performing the evaluation automatically, though they tend to require labeled images to train, which can be time and effort intensive to produce. This work utilizes a semi-supervised training algorithm based on the Meta Pseudo Labels (MPL) model, combined with the feature pyramid network architecture used by AxonDeep, to create a highly adaptable network model that can leverage large amounts of unlabeled data and minimal labeled data to train. This opens up the possibility of being able to quickly retrain the network to adapt to different contexts. This method was applied for semantic segmentations of axons within optic nerve cross-sectional images of mice. The tests performed in this work utilized the same network architecture, training data, and post-processing as an existing deep-learning approach, AxonDeep, to establish a fair comparison. The evaluations performed involved training four models using 10%, 25%, 50%, and 100% of the labeled images (n = 26) alongside unlabeled images (n = 50). Results from the test set (n = 18) show that with 10% of the labeled training data, the MPL model was able to achieve a similar Dice score as the AxonDeep model when trained with 100% of the labeled training data, though the axon-count results calculated during post-processing did not achieve a similar level of accuracy from minimal labeled training data.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Akoglu, H.: User’s guide to correlation coefficients. Turkish J. Emergency Med. 18(3), 91–93 (2018)
Anderson, M.G., et al.: Genetic context determines susceptibility to intraocular pressure elevation in a mouse pigmentary glaucoma. BMC Biol. 4, 1–11 (2006)
Bertels, J., et al.: Optimizing the Dice score and Jaccard index for medical image segmentation: theory and practice. In: Shen, D., et al. (eds.) MICCAI 2019. LNCS, vol. 11765, pp. 92–100. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-32245-8_11
Deng, W., et al.: AxonDeep: automated optic nerve axon segmentation in mice with deep learning. Transl. Vision Sci. Technol. 10(14), 22–22 (2021)
Goyal, V., et al.: AxoNet 2.0: a deep learning-based tool for morphometric analysis of retinal ganglion cell axons. Transl. Vision Sci. Technol. 12(3), 9–9 (2023)
LeCun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature 521(7553), 436–444 (2015)
Mao, M., Hedberg-Buenz, A., Koehn, D., John, S.W., Anderson, M.G.: Anterior segment dysgenesis and early-onset glaucoma in nee mice with mutation of Sh3pxd2b. Invest. Ophthalmol. Visual Sci. 52(5), 2679–2688 (2011)
Pham, H., Dai, Z., Xie, Q., Le, Q.V.: Meta pseudo labels. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 11557–11568 (2021)
Reynaud, J.: Automated quantification of optic nerve axons in primate glaucomatous and normal eyes—method and comparison to semi-automated manual quantification. Investi. Ophthalmol. Visual Sci. 53(6), 2951–2959 (2012)
Ritch, M.D., et al.: AxoNet: a deep learning-based tool to count retinal ganglion cell axons. Sci. Rep. 10(1), 1–13 (2020)
Wang, Y., Zhang, J., Kan, M., Shan, S., Chen, X.: Self-supervised equivariant attention mechanism for weakly supervised semantic segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 12275–12284 (2020)
Zarei, K., et al.: Automated axon counting in rodent optic nerve sections with AxonJ. Sci. Rep. 6(1), 26559 (2016)
Acknowledgements
This study was supported, in part, by the U.S. Department of Veterans Affairs (I50RX003002, I01RX003797, and I01 RX001481), and the National Institutes of Health (T32DK112751, P30 EY025580, and R21 EY029991).
Author information
Authors and Affiliations
Corresponding authors
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
Mann, A., Hedberg-Buenz, A., Anderson, M.G., Garvin, M.K. (2023). Utilizing Meta Pseudo Labels for Semantic Segmentation of Targeted Optic Nerve Features. 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_8
Download citation
DOI: https://doi.org/10.1007/978-3-031-44013-7_8
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)