Abstract
Induced pluripotent stem cells (iPSCs) have broad prospects in clinical and industrial applications, and the automatic identification of iPSCs is essential to optimize the iPSCs manufacturing. However, the challenge of obtaining large amounts of labeled images has limited the use of image analysis in the field of iPSCs. In this paper, we propose Semi-Yolo, a simple yet effective semi-supervised learning (SSL) network that combines mean-teacher SSL paradigm and the Yolo object detection network for the iPSCs identification. The designed burn-in stage in Semi-Yolo makes full use of available labeled images, providing good initialization for both student and teacher model. Then in the teacher-student mutual learning stage, the gradually progressing teacher model of Semi-Yolo generates highly confident pseudo labels for unlabeled images, which provides extra training data to update the student model in a mutually-beneficial manner. Together with a Mosaic data augmentation technique to increase the diversity of our data, Semi-Yolo achieves favorable detection performance on the collected iPSCs detection dataset. Extensive experimental results on our dataset demonstrates the effectiveness of Semi-Yolo with great improvement compared to supervised baseline, which shows better detection precision and faster detection speed than the state-of-the-art SSL object detection algorithm.
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
Takahashi, K., Yamanaka, S.: Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 126(4), 663–676 (2006)
Mahla, R.S.: Stem cells applications in regenerative medicine and disease therapeutics. Int. J. Cell Biol. 2016 (2016)
Zhou, T., Benda, C., Dunzinger, S., et al.: Generation of human induced pluripotent stem cells from urine samples. Nat. Protoc. 7(12), 2080 (2012)
Albarqouni, S., Baur, C., Achilles, F., et al.: AggNet: deep learning from crowds for mitosis detection in breast cancer histology images. IEEE Trans. Med. Imaging 35(5), 1313–1321 (2016)
Li, Z., Dong, M., Wen, S., et al.: CLU-CNNs: object detection for medical images. Neurocomputing 350, 53–59 (2019)
Lee, D.-H.: Pseudo-label: the simple and efficient semi-supervised learning method for deep neural networks. In: Workshop on Challenges in Representation Learning, ICML 2013, vol. 3, no. 2 (2013)
Rasmus, A., Valpola, H., Honkala, M., et al.: Semi-supervised learning with ladder networks. arXiv preprint arXiv:1507.02672 (2015)
Laine, S., Aila, T.: Temporal ensembling for semi-supervised learning. arXiv preprint arXiv:1610.02242 (2016)
Tarvainen, A., Valpola, H.: Mean teachers are better role models: weight-averaged consistency targets improve semi-supervised deep learning results. arXiv preprint arXiv:1703.01780 (2017)
Xie, Q., Luong, M.-T., Hovy, E., et al.: Self-training with noisy student improves imagenet classification. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020, pp. 10687–10698 (2020)
Girshick, R., Donahue, J., Darrell, T., et al.: Rich feature hierarchies for accurate object detection and semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2014, pp. 580–587 (2014)
Girshick, R.: Fast R-CNN. In: Proceedings of the IEEE International Conference on Computer Vision, 2015, pp. 1440–1448 (2015)
Ren, S., He, K., Girshick, R., et al.: Faster R-CNN: towards real-time object detection with region proposal networks. Adv. Neural. Inf. Process. Syst. 2015, 91–99 (2015)
Redmon, J., Farhadi, A.: Yolov3: an incremental improvement. arXiv preprint arXiv:1804.02767 (2018)
Bochkovskiy, A., Wang, C.-Y., Liao, H.-Y.M.: YOLOv4: optimal speed and accuracy of object detection. arXiv preprint arXiv:2004.10934 (2020)
Jocher, G.: Yolo network from ultralytics LLC. https://github.com/ultralytics/yolov5. Accessed 29 Apr 2020
RoyChowdhury, A., Chakrabarty, P., Singh, A., et al.: Automatic adaptation of object detectors to new domains using self-training. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019, pp. 780–790 (2019)
Jeong, J., Lee, S., Kim, J., et al.: Consistency-based semi-supervised learning for object detection (2019)
Sohn, K., Zhang, Z., Li, C.-L., et al.: A simple semi-supervised learning framework for object detection. arXiv preprint arXiv:2005.04757 (2020)
Liu, Y.-C., Ma, C.-Y., He, Z., et al.: Unbiased teacher for semi-supervised object detection. arXiv preprint arXiv:2102.09480 (2021)
Tokunaga, K., Saitoh, N., Goldberg, I.G., et al.: Computational image analysis of colony and nuclear morphology to evaluate human induced pluripotent stem cells. Sci. Rep. 4, 6996 (2014)
Joutsijoki, H., Haponen, M., Rasku, J., et al.: Machine learning approach to automated quality identification of human induced pluripotent stem cell colony images. In: Computational and mathematical methods in medicine, 2016 (2016)
Kavitha, M.S., Kurita, T., Park, S.-Y., et al.: Deep vector-based convolutional neural network approach for automatic recognition of colonies of induced pluripotent stem cells. PloS ONE 12(12) (2017)
Waisman, A., La Greca, A., Möbbs, A.M., et al.: Deep learning neural networks highly predict very early onset of pluripotent stem cell differentiation. Stem Cell Reports 12(4), 845–859 (2019)
Liu, G., David, B.T., Trawczynski, M., et al.: Advances in pluripotent stem cells: history, mechanisms, technologies, and applications. Stem Cell Rev. Rep. 16(1), 3–32 (2020)
Lin, T.-Y., Goyal, P., Girshick, R., et al.: Focal loss for dense object detection. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition 2017, pp. 2980–2988 (2017)
Acknowledgment
This work was supported partly by National Natural Science Foundation of China (Nos. 61871274, U1909209, and No. 62001302), Key Laboratory of Medical Image Processing of Guangdong Province (No. K217300003), Guangdong Pearl River Talents Plan (2016ZT06S220), Guangdong Basic and Applied Basic Research Foundation (Nos. 2021A1515011348, 2019A1515111205), Shenzhen Peacock Plan (Nos. KQTD2016 053112051497 and KQTD2015033016104926), and Shenzhen Key Basic Research Project (JCYJ20170818094109846), Natural Science Foundation of Shenzhen (Nos. JCYJ20190808145011259, RCBS20200714114920379).
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
Wang, X. et al. (2021). Semi-supervised Yolo Network for Induced Pluripotent Stem Cells Detection. In: Peng, Y., Hu, SM., Gabbouj, M., Zhou, K., Elad, M., Xu, K. (eds) Image and Graphics. ICIG 2021. Lecture Notes in Computer Science(), vol 12889. Springer, Cham. https://doi.org/10.1007/978-3-030-87358-5_65
Download citation
DOI: https://doi.org/10.1007/978-3-030-87358-5_65
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-87357-8
Online ISBN: 978-3-030-87358-5
eBook Packages: Computer ScienceComputer Science (R0)