发表机构
Royal Military College of Canada; National Engineering School of Sousse（加拿大皇家军事学院; 苏塞国家工程学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文在乳腺超声、BreakHis 400X 两个数据集上对比七种深度学习模型的能效与性能，发现不同模型在两个数据集上的表现不同，需综合多维度选择医学应用模型。
AI 中文摘要
机器学习领域的最新进展极大提升了乳腺癌检测的准确性与及时性，深度学习（DL）模型在医学图像分析中展现出巨大潜力；然而，随着其架构复杂度提升，环境影响正成为日益突出的问题。本文针对乳腺癌检测任务，在两个医学数据集（乳腺超声数据集、BreakHis 400X）上对七种深度学习模型展开对比分析，评估的架构涵盖卷积神经网络（CNNs）、Transformer 及混合模型。除性能指标外，还评估了训练与推理阶段的二氧化碳排放量。结果显示，EfficientNet 和 ResNet 始终表现出较强性能，但二氧化碳排放量较高；所选 Transformer（如 DeiT-Tiny）在两个数据集上均有具竞争力的表现，而 DenseNet121 的准确率较低。在乳腺超声数据集上，DeiT 在准确率与能耗间取得最有利的平衡；在 BreakHis 数据集上，ViT 和 Swin 模型取得最佳结果。总体而言，本次评估的架构类别中，没有单一类别在两个选定数据集上始终占据主导地位。研究结果强调，在为医学应用选择模型时，需综合考虑性能、碳排放及数据集特性。
英文摘要
Recent advances in machine learning have greatly improved breast cancer detection, enabling more accurate and timely diagnosis. Deep learning (DL) models show strong potential for medical image analysis; however, as their architectural complexity increases, their environmental impacts are becoming a growing concern. In this paper, we present a comparative analysis of seven DL models for breast cancer detection on two medical datasets: Breast Ultrasound and BreakHis 400X. The evaluated architectures range from Convolutional Neural Networks (CNNs) and transformers to hybrid models. In addition to performance metrics, we assess CO2 emissions during both training and inference. Our results show that EfficientNet and ResNet consistently deliver strong performance, although with higher CO2 emissions. The selected transformers, such as DeiT-Tiny, perform competitively on both datasets, whereas DenseNet121 achieves lower accuracy. On the Breast Ultrasound Dataset, DeiT provides the most favourable balance between accuracy and energy consumption, whereas on the BreakHis dataset, the ViT and Swin models achieve the best results. Overall, our findings indicate that no single architecture category from the evaluated ones consistently dominates across the two selected datasets. Our results highlight the importance of jointly considering performance, emissions, and dataset characteristics when selecting models for medical applications.
CommentsAccepted at ICMLA 2026 (IEEE International Conference on Machine Learning and Applications). Camera-ready version submitted
