Effect of Deep Learning Training Policy on Greenhouse Gas Emissions and Carbon Efficiency for Chest Radiograph Classification
胸部X線画像分類における深層学習トレーニング方針が温室効果ガス排出と炭素効率に及ぼす影響 (AI 翻訳)
Nicholas Dietrich, David McShannon, Merel Huisman, Florence X. Doo, Kate Hanneman
🤖 gxceed AI 要約
日本語
本研究は、胸部X線画像分類のための深層学習モデルのトレーニング方針がCO2排出量と性能に与える影響を定量化した。早期打ち切り法は、固定エポック法と同等の性能を維持しつつ、排出量を最大38%削減し、炭素効率を最大76%向上させた。最適チェックポイント後に全排出量の最大84%が発生することが判明した。
English
This study quantified the impact of deep learning training policies on CO2 emissions and performance for chest radiograph classification. Prospective early stopping preserved performance while reducing emissions by up to 38% and improving carbon efficiency by up to 76% compared to fixed-epoch training. Up to 84% of training emissions accrued after the optimal checkpoint.
Unofficial AI-generated summary based on the public title and abstract. Not an official translation.
📝 gxceed 編集解説 — Why this matters
日本のGX文脈において
日本では医療AIの導入が進むが、環境負荷への配慮はまだ少ない。本研究成果は、AI開発における省エネ手法として、医療機関やAIベンダーが持続可能性を考慮する際の参考となる。
In the global GX context
Globally, the environmental footprint of AI is gaining attention. This study provides empirical evidence that simple training policy changes can significantly reduce emissions, contributing to sustainable AI practices in healthcare and beyond.
👥 読者別の含意
🔬研究者:AIトレーニングの炭素排出削減に関する実証データを提供し、効率的なトレーニング戦略の研究に有用。
🏢実務担当者:AIモデル開発時の早期打ち切り採用で、コスト削減と環境負荷低減が可能。
🏛政策担当者:AIの環境影響評価やグリーンAI推進の政策立案に参考となる。
📄 Abstract（原文）
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.
🔗 Provenance — このレコードを発見したソース
- crossref https://doi.org/10.1177/08465371261475319first seen 2026-08-13 05:56:01
🔔 こうした論文の新着を逃したくない方は キーワードアラート に登録（無料・3キーワードまで）。
gxceed は公開メタデータに基づく研究支援データセットです。要約・翻訳・解説は AI 支援で生成されています。 最終的な解釈・検証は利用者が原典資料に基づいて行うことを前提とします。