The ecological footprint of medical AI
Open accessAbstract
Artificial intelligence (AI) is progressively being woven into the fabric of clinical medicine.Most notably, radiology, as an image-based medical specialty, offers numerous applications of AI, including image reconstruction, image analysis, and clinical decision-making.A rich body of research shows the potential of AI in radiology workflows, and indeed, many AI-based medical devices are already available and have received regulatory approval in major markets for clinical routine use.Existing AI models can support radiologists in a wide range of tasks, reaching from image acquisition [1] to diagnosis [2] and outcome prediction [3].In the future, the development of foundation models, i.e., large multimodal transformer neural networks that can manipulate text, images, and other data types, will conceivably increase the use of AI in radiology and medicine even more [4].The advancements in large language models (LLMs), e.g., Chat-GPT, are a testament to these technical developments, as these LLMs already have the capability to reason about clinically relevant topics across a wide range of domains.Behind this effectiveness of contemporaneous AI models are two primary resources: access to large high-quality datasets and the ability to train huge neural networks with up to hundreds of billions of parameters.As these neural networks are trained and are being deployed at scale, a high computational demand is exerted, which consumes electricity and hence leaves a carbon footprint.With the alarming verdict from the Intergovernmental Panel on Climate Change (IPCC-2021) suggesting that
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Smaller models incur smaller carbon footprint and offer reduced serving costs and latency.
“Smaller models incur smaller carbon footprint12 and offer reduced serving costs and latency, which is of particular importance in resource-constrained settings outside of data centers13.”
From A clinically accessible small multimodal radiology model and evaluation metric for chest X-ray findings (2025) · cites this paper as [12]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Foundation models for generalist medical artificial intelligence
Moor, M., Banerjee, O., Abad, Z. S. H., Krumholz, H. M., Leskovec, J., Topol, E. J., & Rajpurkar, P. (2023) · Nature · doi:10.1038/s41586-023-05881-4 · Cited by 2,028 · cited 1× · see citing works
“In the future, the development of foundation models, i.e., large multi-modal transformer neural networks that can manipulate text, images, and other data types, will conceivably increase the use of AI in radiology and medicine even more [4].”
BackgroundAn MRI deep learning model predicts outcome in rectal cancer
Jiang, X., Zhao, H., Saldanha, O. L., Nebelung, S., Kuhl, C., Amygdalos, I., Lang, S. A., Wu, X., Meng, X., Truhn, D., et al. (2023) · Radiology · doi:10.1148/radiol.222223 · Cited by 127 · see citing works
Self-citationEcodesign and Operational Strategies to Reduce the Carbon Footprint of MRI for Energy Cost Savings
Woolen, S. A., Becker, A. E., Martin, A. J., Knoerl, R., Lam, V., Folsom, J., Eusemann, C. D., Hess, C. P., & Deshpande, V. S. (2023) · Radiology · doi:10.1148/radiol.230441 · Cited by 101 · see citing works
Kaniewska, M., Deininger‐Czermak, E., Getzmann, J. M., Wang, X., Lohézic, M., & Guggenberger, R. (2022) · European Radiology · doi:10.1007/s00330-022-09151-1 · Cited by 55 · see citing works
Artificial Intelligence for Clinical Interpretation of Bedside Chest Radiographs
Khader, F., Han, T., Müller‐Franzes, G. A., Huck, L. C., Schad, P., Keil, S., Barzakova, E., Schulze-Hagen, M. F., Pedersoli, F., Schulz, V., et al. (2022) · Radiology · doi:10.1148/radiol.220510 · Cited by 33 · see citing works
Self-citation
Flagged first, then by citations
1 of the 6 references couldn’t be matched to a known paper.
Research works printed but not linked: 3
- [8]Luccioni AS, Viguier S, Ligozat A-L (2022) Estimating the carbon footprint of BLOOM, a 176B parameter language model. https://arxiv.org/abs/2211.02001 · arXiv
- [9]Lacoste A, Luccioni A, Schmidt V, Dandres T (2019) Quantifying the carbon emissions of machine learning. https://arxiv.org/abs/1910.09700 · arXiv
- [10]Patterson D, Gonzalez J, Le Q et al (2021) Carbon emissions and large neural network training. https://arxiv.org/abs/2104.10350 · arXiv
Cited by
Sorted by citations
Artificial intelligence in liver cancer — new tools for research and patient management
Caldéraro, J., Žigutytė, L., Truhn, D., Jaffe, A., & Kather, J. N. (2024) · Nature Reviews Gastroenterology & Hepatology · doi:10.1038/s41575-024-00919-y · Cited by 72 · see citing works
Self-citationZambrano Chaves, J. M., Huang, S.-C., Xu, Y., Xu, H., Usuyama, N., Zhang, S., Wang, F., Xie, Y., Khademi, M., Yang, Z., et al. (2025) · Nature Communications · doi:10.1038/s41467-025-58344-x · Cited by 45 · see citing works
“Smaller models incur smaller carbon footprint 12 and offer reduced serving costs and latency, which is of particular importance in resource-constrained settings outside of data centers 13 .”
Best Practices for Large Language Models in Radiology
Blüthgen, C., Veen, D. V., Zakka, C., Link, K. E., Fanous, A. H., Daneshjou, R., Frauenfelder, T., Langlotz, C. P., Gatidis, S., & Chaudhari, A. S. (2025) · Radiology · doi:10.1148/radiol.240528 · Cited by 41 · see citing works
The barriers to uptake of artificial intelligence in hepatology and how to overcome them
Clusmann, J., Balaguer‐Montero, M., Bassegoda, O., Schneider, C. V., Seraphin, T. P., Paintsil, E. K., Luedde, T., Pérez-López, R., Caldéraro, J., Gilbert, S., et al. (2025) · Journal of Hepatology · doi:10.1016/j.jhep.2025.07.003 · Cited by 29 · see citing works
Self-citationRetractedAgentic Large Language Models for Healthcare: Current Progress and Future Opportunities
Yuan, H. (2025) · Medicine Advances · doi:10.1002/med4.70000 · Cited by 23 · see citing works
The green ICU: how to interpret green? A multiple perspective approach
Smale, E. M., Baid, H., Balan, M., McGain, F., McAlistar, S., Waele, J. J. D., Diehl, J. C., van Raaij, E. M., van Genderen, M. E., Tibboel, D., et al. (2025) · Critical Care · doi:10.1186/s13054-025-05316-8 · Cited by 27 · see citing works
A Systematic Review and Evaluation of Sustainable AI Algorithms and Techniques in Healthcare
Alzoubi, Y. I., Topcu, A. E., & Α, E. E. (2025) · IEEE Access · doi:10.1109/access.2025.3596189 · Cited by 15 · see citing works
Canadian Association of Radiologists Statement on Planetary Health Education in Radiology
Brown, M. J., Forster, B. B., McInnes, M. D. F., Komar, M. A., Amin, P., Atwal, S., Chen, S., Hamwi, M., Ladak, R., Malik, A., et al. (2024) · Canadian Association of Radiologists Journal · doi:10.1177/08465371241279359 · Cited by 15 · see citing works
Artificial intelligence, sustainability and environmental impact. A narrative and bibliometric study
Camastra, F. D., & Vallejo, R. G. (2025) · Región Científica · doi:10.58763/rc2025355 · Cited by 20 · see citing works
Estimation of carbon footprint in nuclear medicine: illustration of a french department
Godard, F., Oosthoek, J., Alexis, A., Leo, P., Fontaine, E., Dahmani, M., Houot, L., Quermonne, M., Cochet, A., & Drouet, C. (2025) · European Journal of Nuclear Medicine and Molecular Imaging · doi:10.1007/s00259-025-07129-x · Cited by 9 · see citing works
Jia, Y., Deng, M., Burger, R., Sheard, S., Hanneman, K. A., Iarovich, M. D., Sala, E., Avesani, G., Illing, R. O., & Rockall, A. (2025) · European Radiology · doi:10.1007/s00330-025-12023-z · Cited by 6 · see citing works
Exploring environmental sustainability of artificial intelligence in radiology: A scoping review
Champendal, M., Lokaj, B., de Gevigney, V. D., Brulé, G., Zaghir, J., Boiko, P., Lovis, C., Müller, H., Schmid, J., & Ribeiro, R. T. (2025) · European Journal of Radiology · doi:10.1016/j.ejrad.2025.112558 · Cited by 6 · see citing works
Liu, Q., Li, Y., Hao, Y., Fan, W., Liu, J., Li, T., & Liu, L. (2024) · Frontiers in Oncology · doi:10.3389/fonc.2024.1349388 · Cited by 6 · see citing works
Carbon Reporting Practices in the NHS: Emissions and Omissions Relating to Artificial Intelligence
Reynolds, D. J. (2025) · Journal of Medical Internet Research · doi:10.2196/79174 · Cited by 5 · see citing works
The AI Efficiency Paradox: Reclaiming Quality Patient Care in an Era of Optimization
Burwell, J. M. (2025) · Journal of Medical Systems · doi:10.1007/s10916-025-02183-2 · Cited by 5 · see citing works
Artificial intelligence in public health—challenges and opportunities
Raj, T. D. S., Philip, A., & Thomas, V. (2026) · European Journal of Clinical Nutrition · doi:10.1038/s41430-026-01745-y · Cited by 5 · see citing works
Ramachandran, A., Sarabu, C. R., Gupta, U., Ghose, S., & Lee, V. S. (2025) · NEJM Catalyst · doi:10.1056/cat.25.0125 · Cited by 3 · see citing works
Karageorgos, S. A., Hibberd, O., Ren, D., Hornsby, Y., Roland, D., & Koutroulis, I. (2025) · Children · doi:10.3390/children12111542 · Cited by 2 · see citing works
A comprehensive review of efficient deep learning for clinical medical imaging deployment
Nguyen, C. M., & Hy, T. S. (2026) · Discover Artificial Intelligence · doi:10.1007/s44163-026-01706-1 · Cited by 2 · see citing works
Bridging the Bench-to-Bedside Gap with Multimodal Artificial Intelligence in Digestive Diseases
Weiskirchen, R. (2026) · Livers · doi:10.3390/livers6010001 · Cited by 1 · see citing works
Hosni, E., & Jemai, A. (2025) · IEEE Afro-Mediterranean Conference on Artificial Intelligence (AMCAI) · doi:10.1109/amcai66110.2025.11474358 · Cited by 1 · see citing works
Clinical Implementation of Artificial Intelligence Systems in Medical Imaging and Radiotherapy
Vienna International Centre, 100, P. B., Vienna, A.-1., Austria, 2600-0, T. +. (., & 2600-7, F. +. (. (2026) · IAEA human health series · doi:10.61092/iaea.9olq-aqa9 · Cited by 2 · see citing works
Offene Fragen und grundsätzliche Probleme
Kather, J. N. (2025) · Springer eBooks · doi:10.1007/978-3-662-71042-5_7
Self-citationLei, I. I., Arasaradnam, R. P., Marlicz, W. M., & Koulaouzidis, A. (2025) · Diagnostics · doi:10.3390/diagnostics15232938
Ethics in Artificial Intelligence and Precision Medicine: A Perspective
Seyhan, A. A., & Carini, C. (2025) · Springer eBooks · doi:10.1007/978-3-032-07649-6_45
Ethics Applied to Digital Healthcare: Navigating the Moral Landscape
Niazkhani, Z., Wallenburg, I., Hendriks, J. M., & Wehrens, R. (2026) · Springer eBooks · doi:10.1007/978-3-031-89820-4_24
Zakrison, T. L., Poulson, M. R., & Richardson, J. (2026) · Journal of the American College of Surgeons · doi:10.1097/xcs.0000000000001951
Introduction: Sustainability in Radiology and Healthcare
Ranschaert, E. R., Klontzas, M. E., & Safdar, N. M. (2026) · Imaging informatics for healthcare professionals · doi:10.1007/978-3-032-15693-8_1
Artificial intelligence in healthcare: concepts and applications
Das, S., Chanda, P. B., & Maity, D. (2026) · Elsevier eBooks · doi:10.1016/b978-0-443-45875-0.00008-9
Gauss, T., Delude, T. F., Kalimouttou, A., Seddiki, O., Sanchez, C., Grèze, J., Brossard, C., Moyer, J.-D., Brelurut, G., Medjkoune, S., et al. (2026) · medRxiv · doi:10.64898/2026.07.05.26357337
Environmental impact of portable MRI in remote Canadian settings
Azizi, N., Zhou, Z., Ibrahim, A., & Islam, O. (2026) · Frontiers in Neuroimaging · doi:10.3389/fnimg.2026.1706184
Kumar, K., & Jyoti, K. (2026) · Results in Engineering · doi:10.1016/j.rineng.2026.112746
- Authors
- Truhn, D., Müller-Franzes, G., & Kather, J. N.
- Affiliations
- Universitätsklinikum Aachen (DE)
- University Hospital Heidelberg (DE)
- Corresponding author
- Truhn, Daniel
- Venue
- European Radiology
- Published
- 2023-08-15
- Volume
- 34
- Issue
- 2
- Pages
- 1176-1178
- ISSN
- 0938-7994 (ISSN-L)
- 1432-1084
- PMC
- PMC10853292
- OpenAlex
- W4385799552
- Cited by
- 32
- References
- 6
- Citations by year
- 2026: 11 · 2025: 18 · 2024: 3
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Locations
- European Radiology (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- RWTH Publications (RWTH Aachen) (publishedVersion, other-oa)
- PUBLISSO (German National Library of Medicine) (submittedVersion, cc-by)
- RWTH Publications (RWTH Aachen)
- Funders
- RWTH Aachen University
- Topics
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Green IT and Sustainability
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Physical Sciences › Environmental Science › Environmental Engineering › Environmental Impact and Sustainability
- Keywords
- ecological footprint, artificial intelligence, carbon footprint, radiology, large language models, foundation models, large-scale neural networks, energy consumption, artificial intelligence in healthcare
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
