堆叠集成与ECA-EfficientNetV2卷积神经网络对包括COVID-19在内的多种胸部疾病分类
本研究提出Stacking-ensemble和ECA-EfficientNetV2两种模型,在胸部X线和CT图像上对多种胸部疾病(包括COVID-19)进行分类,采用十折交叉验证。ECA-EfficientNetV2在X线数据集上准确率99.21%、AUC 99.51%,在CT数据集上准确率99.81%、AUC 99.87%。集成模型性能优于单一预训练模型,两种模型间五项指标差异不显著。
要点
- ECA-EfficientNetV2在胸部X线和CT多疾病分类中均表现优异
- 集成模型性能优于单一预训练模型,但差异不显著
- 研究为回顾性算法开发,尚未验证临床实用性
核心结论
- 集成模型性能优于单一预训练模型,ECA-EfficientNetV2与Stacking-ensemble在多种胸部疾病分类上均表现良好。
研究设计与样本
- 研究为算法开发与验证,提出Stacking-ensemble和ECA-EfficientNetV2两种模型,在胸部X线和CT图像上进行十折交叉验证。
- Stacking-ensemble堆叠了EfficientNetV2-B0、B1、B2、B3、S和M六种预训练模型;ECA-EfficientNetV2基于ECA-Net和EfficientNetV2自行设计。
- 另使用COVID-CT数据集验证模型性能。
诊断标准与关键数值
- ECA-EfficientNetV2在胸部X线数据集上准确率99.21%、精确率99.23%、召回率99.25%、F1分数99.20%、AUC 99.51%。
- ECA-EfficientNetV2在胸部CT数据集上准确率99.81%、精确率99.80%、召回率99.80%、F1分数99.81%、AUC 99.87%。
- Stacking-ensemble与ECA-EfficientNetV2在五项指标上的差异不显著。
临床应用与阅片要点
- 研究提示深度学习模型可辅助胸部X线和CT上多种胸部疾病(包括COVID-19)的分类,但尚未在临床工作流中验证。
- 阅片时需注意模型分类结果不能替代放射科医师的综合诊断。
局限性
- 原文未明确报告研究设计的具体局限,但研究为算法开发,未涉及临床落地评估。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
RATIONALE AND OBJECTIVES: Early detection and treatment of COVID-19 patients is crucial. Convolutional neural networks have been proven to accurately extract features in medical images, which accelerates time required for testing and increases the effectiveness of COVID-19 diagnosis. This study proposes two classification models for multiple chest diseases including COVID-19.
MATERIALS AND METHODS: The first is Stacking-ensemble model, which stacks six pretrained models including EfficientNetV2-B0, EfficientNetV2-B1, EfficientNetV2-B2, EfficientNetV2-B3, EfficientNetV2-S and EfficientNetV2-M. The second model is self-designed model ECA-EfficientNetV2 based on ECA-Net and EfficientNetV2. Ten-fold cross validation was performed for each model on chest X-ray and CT images. One more dataset, COVID-CT dataset, was tested to verify the performance of the proposed Stacking-ensemble and ECA-EfficientNetV2 models.
RESULTS: The best performance comes from the proposed ECA-EfficientNetV2 model with the highest Accuracy of 99.21%, Precision of 99.23%, Recall of 99.25%, F1-score of 99.20%, and (area under the curve) AUC of 99.51% on chest X-ray dataset; the best performance comes from the proposed ECA-EfficientNetV2 model with the highest Accuracy of 99.81%, Precision of 99.80%, Recall of 99.80%, F1-score of 99.81%, and AUC of 99.87% on chest CT dataset. The differences for five metrics between Stacking-ensemble and ECA-EfficientNetV2 models are not significant.
CONCLUSION: Ensemble model achieves better performance than single pretrained models. Compared to the SOTA, Stacking-ensemble and ECA-EfficientNetV2 models proposed in this study demonstrate promising performance on classification of multiple chest diseases including COVID-19.
引用
- Huang, Liao. Stacking Ensemble and ECA-EfficientNetV2 Convolutional Neural Networks on Classification of Multiple Chest Diseases Including COVID-19. Acad Radiol. 2022. PMID: 36526533. doi:10.1016/j.acra.2022.11.027
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。