基于深度学习的儿童肺炎病原学类型判别模型
|
摘要
目的:构建儿童肺炎病原学类型自动判别模型,从临床上规范儿童肺炎治疗用药。方法:利用深度学习模型,结合迁移学习技术,对肺炎胸片首先进行肺区域分割,其次以痰培养结果作为金标准,对肺区域进行病原学类型为病毒或细菌的判别。结果:基于深度卷积神经网络的肺炎病原学类型二分类判别模型的准确率达80.48%,特异度82.07%,灵敏度77.55%,AUC达0.82。结论:基于深度学习技术和胸片数据的肺炎病原学类型判别模型,能够对肺炎治疗用药提供辅助决策支持,降低试药风险,使患者及早得到治疗。
Objective: To construct an automatic classification model for pathogens and thus regulate the treatment of pediatric pneumonia. Methods: By the deep learning model, combined with transfer learning technology, lung regions were segmented from the X-ray of pneumonia patients. Then the model classified the lung regions as viral or bacterial, with the results of sputum culture as ground truths. Results: The binary classification model for the pathogen of pneumonia based on deep convolutional neural network can achieve an accuracy of 80.48%, specificity of 82.07%, sensitivity of 77.55% and AUC of 0.82. Conclusion: The classification model for pneumonia pathogens based on X-ray data and deep learning methods can provide clinical decision support during for treatment,decrease risks and make patients receive timely treatment.
Objective: To construct an automatic classification model for pathogens and thus regulate the treatment of pediatric pneumonia. Methods: By the deep learning model, combined with transfer learning technology, lung regions were segmented from the X-ray of pneumonia patients. Then the model classified the lung regions as viral or bacterial, with the results of sputum culture as ground truths. Results: The binary classification model for the pathogen of pneumonia based on deep convolutional neural network can achieve an accuracy of 80.48%, specificity of 82.07%, sensitivity of 77.55% and AUC of 0.82. Conclusion: The classification model for pneumonia pathogens based on X-ray data and deep learning methods can provide clinical decision support during for treatment,decrease risks and make patients receive timely treatment.
引文
[1]Chartrand G,Cheng PM,Vorontsov E,et al.Deep Learning:A Primer for Radiologists[J].Radiographics,2017,37(7):2113-2131.
[2]Litjens G,Kooi T,Bejnordi BE,et al.A survey on deep learning in medical image analysis[J].Med Image Anal,2017(42):60-88.
[3]Gulshan V,Peng L,Coram M,et al.Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs[J].JAMA,2016,316(22):2402-2410.
[4]Rajpurkar P,Irvin J,Zhu K,et al.Chexnet:Radiologist-level pneumonia detection on chest x-rays with deep learning[J].Ar Xiv,2017:1711.05225.
[5]Long J,Shelhamer E,Darrell T.Fully convolutional networks for semantic segmentation[C]//Proceedings of the IEEEconference on computer vision and pattern recognition,2015:3431-3440.
[6]Pan SJ,Yang Q.A survey on transfer learning[J].IEEE Transactions on knowledge and data engineering,2010,22(10):1345-1359.
[2]Litjens G,Kooi T,Bejnordi BE,et al.A survey on deep learning in medical image analysis[J].Med Image Anal,2017(42):60-88.
[3]Gulshan V,Peng L,Coram M,et al.Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs[J].JAMA,2016,316(22):2402-2410.
[4]Rajpurkar P,Irvin J,Zhu K,et al.Chexnet:Radiologist-level pneumonia detection on chest x-rays with deep learning[J].Ar Xiv,2017:1711.05225.
[5]Long J,Shelhamer E,Darrell T.Fully convolutional networks for semantic segmentation[C]//Proceedings of the IEEEconference on computer vision and pattern recognition,2015:3431-3440.
[6]Pan SJ,Yang Q.A survey on transfer learning[J].IEEE Transactions on knowledge and data engineering,2010,22(10):1345-1359.