EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Effect of Deep Learning Training Policy on Greenhouse Gas Emissions and Carbon Efficiency for Chest Radiograph Classification

Nicholas Dietrich iD, David McShannon, Merel Huisman iD, Florence X Doo iD, Kate Hanneman iD

Canadian Association of Radiologists Journal · 12.08.2026

DOI: 10.1177/08465371261475319 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Canadian Association of Radiologists Journal, online erschienen am 12.08.2026. Verfasst von Nicholas Dietrich und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Medical Imaging, Temerty Faculty of Medicine, University of Toronto, ON, Canada. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: University of Maryland, Baltimore, Institute for Clinical & Translational Research, Montgomery County, Maryland, University of Maryland Strategic Partnership: MPowering the State.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

Environmental sustainability is an emerging priority in radiology, yet the impact of deep learning training policies on greenhouse gas emissions remains poorly characterized. This study quantified the effect of training policy on carbon dioxide equivalent (CO 2 eq) emissions and model performance for chest radiograph classification.

Methods

Anteroposterior chest radiographs (128 907 training, 24 570 validation, 8282 test) were used to train 3 ImageNet-pretrained convolutional neural networks (ResNet-50, DenseNet-121, EfficientNet-B0) for 20 epochs. Three policies were evaluated: (1) retrospective optimal checkpoint selection at the validation loss minimum; (2) prospective early stopping (patience 10 epochs); and (3) fixed 20-epoch training without checkpoint selection. Per-epoch CO 2 eq emissions, energy, macro-averaged area under the curve (AUC), and carbon efficiency were evaluated.

Results

Validation loss reached its minimum at median epoch 2 for ResNet-50 and DenseNet-121 and epoch 4 for EfficientNet-B0. At the retrospective optimum, macro-AUCs ranged from 0.793 to 0.800 and generated 6.2 to 7.9 g CO 2 eq (37-46 Wh) at the deployed checkpoint. However, producing this model required the full run, generating 30.8 to 49.4 g CO 2 eq (181-291 Wh) with 78% to 84% of training emissions accruing after the deployed checkpoint. Prospective early stopping had macro-AUCs equivalent to the retrospective optimum (0.793-0.800), with 31% to 38% lower total emissions (21.1-30.9 vs 30.8-49.4 g CO 2 eq) and 57% to 76% higher carbon efficiency (25.9-37.6 vs 14.7-24.0 AUC/kg CO 2 eq) compared to fixed-epoch training.

Conclusions

Up to 84% of total training emissions accrued after the optimal checkpoint, with relative savings dependent on the comparator. Prospective early stopping preserved performance, reduced emissions by up to 38%, and improved carbon efficiency by up to 76% versus fixed 20-epoch training.

Schlagwörter

Stichwörter der Autor:innen

artificial intelligencecarbon efficiencychest radiographdeep learningenvironmental sustainabilitygreenhouse gas emissionstraining policy

Autor:innen und Institutionen

  • Department of Medical Imaging, Temerty Faculty of Medicine, University of Toronto, ON, Canada.
  • Faculty of Engineering, McMaster University, Hamilton, ON, Canada.
  • Department of Radiology and Nuclear Medicine, Radboud University Medical Center, Nijmegen, The Netherlands.
  • University of Maryland Medical Intelligent Imaging (UM2ii) Center, Department of Radiology and Nuclear Medicine, University of Maryland, Baltimore, MD, USA.
  • Center for Applied AI, University of Maryland-Institute for Health Computing (UM-IHC), North Bethesda, MD, USA.
  • Joint Department of Medical Imaging, University Medical Imaging Toronto, University Health Network, ON, Canada.

ORCID-Profile: Nicholas Dietrich, Merel Huisman, Florence X Doo, Kate Hanneman

Förderung

  • University of Maryland, Baltimore, Institute for Clinical & Translational Research
  • Montgomery County, Maryland
  • University of Maryland Strategic Partnership: MPowering the State
  • NVIDIA
  • Microsoft Azure

Zitieren

APA

Dietrich, N., McShannon, D., Huisman, M., Doo, F. X., & Hanneman, K. (2026). Effect of Deep Learning Training Policy on Greenhouse Gas Emissions and Carbon Efficiency for Chest Radiograph Classification. Canadian Association of Radiologists Journal. https://doi.org/10.1177/08465371261475319

Vancouver

Dietrich N, McShannon D, Huisman M, Doo FX, Hanneman K. Effect of Deep Learning Training Policy on Greenhouse Gas Emissions and Carbon Efficiency for Chest Radiograph Classification. Canadian Association of Radiologists Journal. 2026. doi:10.1177/08465371261475319

BibTeX
@article{Dietrich2026,
  author = {Dietrich, Nicholas and McShannon, David and Huisman, Merel and Doo, Florence X and Hanneman, Kate},
  title = {Effect of Deep Learning Training Policy on Greenhouse Gas Emissions and Carbon Efficiency for Chest Radiograph Classification},
  journal = {Canadian Association of Radiologists Journal},
  year = {2026},
  doi = {10.1177/08465371261475319}
}
Kostenlos starten →