Abstract
We present how replacing convolutional neural networks with a rotation-equivariant counterpart can be used to reduce the amount of training images needed for classification of whether a cell is cancerous or not. Our hypothesis is that data augmentation schemes by rotation can be replaced, thereby increasing weight sharing and reducing overfitting. The dataset at hand consists of single cell images. We have balanced a subset of almost 9.000 images from healthy patients and patients diagnosed with cancer. Results show that classification accuracy is improved and overfitting reduced if compared to an ordinary convolutional neural network. The results are encouraging and thereby an advancing step towards making screening of patients widely used for the application of oral cancer.
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
Similar content being viewed by others
References
Bengtsson, E., Malm, P.: Computational and mathematical methods in medicine. J. Big Data (2014)
Chidester, B., Zhou, T., Do, M.N., Ma, J.: Rotation equivariant and invariant neural networks for microscopy image analysis. Bioinformatics 35(14), 530–537 (2019)
Cohen, T., Welling, M.: Group equivariant convolutional networks. In: International Conference on Machine Learning (ICML) (2016)
Forslid, G., et al.: Deep convolutional neural networks for detecting cellular changes due to malignancy. In: International Conference on Computer Vision Workshops (ICCVW) (2017)
Galbraith, W., Marshall, P., Lee, E., Bacus, J.: Studies on Papanicolaou staining. I. Visible-light spectra of stained cervical cells. Anal. Quant. Cytol. 1(3), 160–168 (1979)
Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning, p. 127. MIT Press, Cambridge (2016)
Holmquist, J., Imasoto, Y., Bengtsson, E., Olsen, B., Stenkvist, B.: A microspectrophotometric study of Papanicolaou-stained cervical cells as an aid in computerized image processing. J. Histochem. Cytochem. 24(12), 1218–1224 (1976)
Lu, J., Sladoje, N., Stark Runow, C., Darai Ramqvist, E., Hirsch, J.M., Lindblad, J.: A deep learning based pipeline for efficient oral cancer screening on whole slide images. In: International Conference on Image Analysis and Recognition (ICIAR) (2020)
Semih Kayhan, O., van Gemert, J.C.: On translation invariance in CNNs: convolutional layers can exploit absolute spatial location. In: Conference on Computer Vision and Pattern Recognition (CVPR) (2020)
Shorten, C., Khoshgoftaar, T.: A survey on image data augmentation for deep learning. J. Big Data 6, 60 (2019)
Vapnik, V.: Principles of risk minimization for learning theory. In: Moody, J.E., Hanson, S.J., Lippmann, R.P. (eds.) Advances in Neural Information Processing Systems, vol. 4, pp. 831–838. Morgan-Kaufmann (1992)
Weiler, M., Cesa, G.: General E(2)-Equivariant Steerable CNNs. In: Conference on Neural Information Processing Systems (NeurIPS) (2019)
Weiler, M., Hamprecht, F., Storath, M.: Learning steerable filters for rotation equivariant CNNs. In: Conference on Computer Vision and Pattern Recognition (CVPR) (2018)
Wetzer, E., Gay, J., Harlin, H., Lindblad, J., Sladoje, N.: When texture matters: texture-focused CNNs outperform general data augmentation and pretraining in oral cancer detection. In: International Symposium on Biomedical Imaging (ISBI) (2020)
Winkels, M., Cohen, T.: 3D G-CNNs for pulmonary nodule detection. In: International Conference on Medical Imaging with Deep Learning (2018)
Worrall, D., Brostow, G.: CubeNet: equivariance to 3D rotation and translation. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11209, pp. 585–602. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01228-1_35
Acknowledgment
We thank Gabriele Cesa et al. for their contributions during our tailoring of the framework to the specific application of cancer classification for oral images. We are grateful for Professor Emeritus Ewert Bengtsson’s valuable input to this project on cell image analysis in general and color conversions in particular. We especially thank Professor Nataša Sladoje for initiating the project in conjunction with the Wallenberg AI, Autonomous Systems and Software Program.
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Bernander, K.B., Lindblad, J., Strand, R., Nyström, I. (2021). Replacing Data Augmentation with Rotation-Equivariant CNNs in Image-Based Classification of Oral Cancer. In: Tavares, J.M.R.S., Papa, J.P., González Hidalgo, M. (eds) Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications. CIARP 2021. Lecture Notes in Computer Science(), vol 12702. Springer, Cham. https://doi.org/10.1007/978-3-030-93420-0_3
Download citation
DOI: https://doi.org/10.1007/978-3-030-93420-0_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-93419-4
Online ISBN: 978-3-030-93420-0
eBook Packages: Computer ScienceComputer Science (R0)
