Abstract
The increasing energy consumption and carbon footprint of deep learning (DL) due to growing compute requirements has become a cause of concern. In this work, we focus on the carbon footprint of developing DL models for medical image analysis (MIA), where volumetric images of high spatial resolution are handled. We present and compare the features of four tools from literature to quantify the carbon footprint of DL. Using one of these tools we estimate the carbon footprint of medical image segmentation pipelines. We choose nnU-net as the proxy for a medical image segmentation pipeline and experiment on three common datasets. With our work we hope to inform on the increasing energy costs incurred by MIA. We discuss simple strategies to cut-down the environmental impact that can make model selection and training processes more efficient.
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.
Underlining by us for emphasis.
- 2.
- 3.
- 4.
- 5.
When the trackers cannot fetch real-time carbon intensity of energy for the specific geographic location, most resort to using some average estimate from a look-up table.
References
Alcott, B.: Jevons’ paradox. Ecol. Econ. 54(1), 9–21 (2005)
Amodei, D., Hernandez, D., Sastry, G., Clark, J., Brockman, G., Sutskever, I.: AI and compute (2018). https://blog.openai.com/aiand-compute
Anthony, L.F.W., Kanding, B., Selvan, R.: Carbontracker: tracking and predicting the carbon footprint of training deep learning models. In: ICML Workshop on Challenges in Deploying and monitoring Machine Learning Systems, July 2020. arXiv:2007.03051
Bender, E.M., Gebru, T., McMillan-Major, A., Shmitchell, S.: On the dangers of stochastic parrots: can language models be too big? In: Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, pp. 610–623 (2021)
Bergstra, J., Bengio, Y.: Random search for hyper-parameter optimization. J. Mach. Learn. Res. 13, 281–305 (2012)
Brock, A., Lim, T., Ritchie, J., Weston, N.: SMASH: one-shot model architecture search through hypernetworks. In: International Conference on Learning Representations (2018). https://openreview.net/forum?id=rydeCEhs-
de Bruijne, M., et al. (eds.): MICCAI 2021, Part III. LNCS, vol. 12903. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-87199-4
Change, I.C., et al.: Mitigation of climate change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, vol. 1454, p. 147 (2014)
Fei-Fei, L., Deng, J., Li, K.: ImageNet: Constructing a large-scale image database. J. Vis. 9(8), 1037 (2009)
Energy, G.: CO2 Status Report—The Latest Trends in Energy and Emissions in 2018. International Energy Agency (2019). https://www.iea.org/reports/global-energy-co2-status-report-2019/emissions
European-Environment-Agency: Greenhouse gas emission intensity of electricity generation by country (2022). https://www.eea.europa.eu/data-and-maps/daviz/co2-emission-intensity-9/
Grealey, J., et al.: The carbon footprint of bioinformatics. BioRxiv (2021)
Henderson, P., Hu, J., Romoff, J., Brunskill, E., Jurafsky, D., Pineau, J.: Towards the systematic reporting of the energy and carbon footprints of machine learning. J. Mach. Learn. Res. 21(248), 1–43 (2020)
Hoopes, A., Hoffmann, M., Fischl, B., Guttag, J., Dalca, A.V.: HyperMorph: amortized hyperparameter learning for image registration. In: Feragen, A., Sommer, S., Schnabel, J., Nielsen, M. (eds.) IPMI 2021. LNCS, vol. 12729, pp. 3–17. Springer, Cham (2021). https://doi.org/10.1007/978-3-030-78191-0_1
Isensee, F., Jaeger, P.F., Kohl, S.A., Petersen, J., Maier-Hein, K.H.: nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation. Nat. Methods 18(2), 203–211 (2021)
Kumar, N., et al.: A multi-organ nucleus segmentation challenge. IEEE Trans. Med. Imaging 39(5), 1380–1391 (2019)
Lacoste, A., Luccioni, A., Schmidt, V., Dandres, T.: Quantifying the carbon emissions of machine learning. Technical report (2019)
Lannelongue, L.: Carbon footprint: the (not so) hidden cost of high performance computing. ITNOW 63(4), 12–13 (2021)
LeCun, Y., Bengio, Y., Hinton, G.: Deep learning. Nature 521(7553), 436–444 (2015)
Masson-Delmotte, V., Zhai, P., Pirani, A., Connors, S., et al. (eds.): Summary for Policymakers. Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change (2021)
Micikevicius, P., et al.: Mixed precision training. In: International Conference on Learning Representations (2018). https://openreview.net/forum?id=r1gs9JgRZ
Paszke, A., et al.: PyTorch: an imperative style, high-performance deep learning library. In: Advances in Neural Information Processing Systems, pp. 8024–8035 (2019)
Rae, C., Farley, M., Jeffery, K., Urai, A.E.: Climate crisis and ecological emergency: why they concern (neuro) scientists, and what we can do (2021)
Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-24574-4_28http://arxiv.org/abs/1505.04597
Samuel, G., Lucivero, F., Lucassen, A.: Sustainable biobanks: a case study for a green global bioethics. Glob. Bioeth. 33(1), 50–64 (2022)
Schmidhuber, J.: Deep learning in neural networks: an overview. Neural Netw. 61, 85–117 (2015)
Schmidt, V., et al.: CodeCarbon: estimate and track carbon emissions from machine learning computing (2021). https://doi.org/10.5281/zenodo.4658424
Selvan, R.: Carbon footprint driven deep learning model selection for medical imaging. In: Medical Imaging with Deep Learning (Short Paper Track) (2021)
Selvan, R., Bhagwat, N., Anthony, L.F.W., Kanding, B., Dam, E.B.: Carbon footprint of selecting and training deep learning models for medical image analysis (supplementary material). arXiv preprint arXiv:2203.02202 (2022)
Sevilla, J., Heim, L., Ho, A., Besiroglu, T., Hobbhahn, M., Villalobos, P.: Compute trends across three eras of machine learning. arXiv preprint arXiv:2202.05924 (2022)
Simpson, A.L., et al.: A large annotated medical image dataset for the development and evaluation of segmentation algorithms. arXiv preprint arXiv:1902.09063 (2019)
Strubell, E., Ganesh, A., McCallum, A.: Energy and policy considerations for deep learning in NLP. In: Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pp. 3645–3650 (2019)
U.S. Energy-Information-Administration: United states electricity profile 2020 (2022). https://www.eia.gov/electricity/state/unitedstates/
Worlbank-DataBank: Sustainable Development Goals (SDGs): CO2 emissions per capita (2022). Data retrieved from World Development Indicators on 28/02/2022. https://databank.worldbank.org/source/sustainable-development-goals-(sdgs)
Yang, T.J., Chen, Y.H., Sze, V.: Designing energy-efficient convolutional neural networks using energy-aware pruning. In: Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017. vol. 2017, pp. 6071–6079, November 2017. https://doi.org/10.1109/CVPR.2017.643. https://arxiv.org/abs/1611.05128
Acknowledgments
The authors would like to thank members of OHBM SEA-SIG community for insightful and thought-provoking discussions on environmental sustainability and MIA.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Selvan, R., Bhagwat, N., Wolff Anthony, L.F., Kanding, B., Dam, E.B. (2022). Carbon Footprint of Selecting and Training Deep Learning Models for Medical Image Analysis. In: Wang, L., Dou, Q., Fletcher, P.T., Speidel, S., Li, S. (eds) Medical Image Computing and Computer Assisted Intervention – MICCAI 2022. MICCAI 2022. Lecture Notes in Computer Science, vol 13435. Springer, Cham. https://doi.org/10.1007/978-3-031-16443-9_49
Download citation
DOI: https://doi.org/10.1007/978-3-031-16443-9_49
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-16442-2
Online ISBN: 978-3-031-16443-9
eBook Packages: Computer ScienceComputer Science (R0)
