基于胸部CT图像的肺炎深度学习分类预测模型

曾梦; 赵娜; 王显棋; 庞慧琳; 侯文静; 刘玲玲; 王晓雯; 龚家利; 曾文兵; 崔洪亮; 陈伟

doi:10.16016/j.2097-0927.202307139

您当前的位置：

首页 >

文章列表页 >

基于胸部CT图像的肺炎深度学习分类预测模型

临床医学 | 更新时间：2026-01-09

- 基于胸部CT图像的肺炎深度学习分类预测模型
- Establishment of a chest-CT-image-based classification deep learning model for pneumonia prediction
- 陆军军医大学学报 2023年45卷第21期页码：2266-2274
- 作者机构：
  
  1.400038 重庆，陆军军医大学(第三军医大学)第一附属医院放射科
  2.401329 重庆，重庆知见生命科技有限公司
  3.404000 重庆，重庆大学附属三峡医院放射科
- 作者简介：
  
  [陈伟，E-mail：landcw@163.comCHEN Wei, E-mail: landcw@163.com
- 基金信息：
  
  陆军军医大学第一附属医院新冠病毒感染临床救治科技攻关应急项目(2023XGIIT06)
- DOI：10.16016/j.2097-0927.202307139
  中图分类号： R319;R563.1;R814.42
- 收稿：2023-07-30，
  
  修回：2023-09-12，
  
  纸质出版：2023-11-15
- 稿件说明：
移动端阅览
曾梦, 赵娜, 王显棋, 等. 基于胸部CT图像的肺炎深度学习分类预测模型[J]. 陆军军医大学学报, 2023,45(21):2266-2274.

Meng ZENG, Na ZHAO, Xianqi WANG, et al. Establishment of a chest-CT-image-based classification deep learning model for pneumonia prediction[J]. Journal of Army Medical University, 2023, 45(21): 2266-2274.
曾梦, 赵娜, 王显棋, 等. 基于胸部CT图像的肺炎深度学习分类预测模型[J]. 陆军军医大学学报, 2023,45(21):2266-2274. DOI： 10.16016/j.2097-0927.202307139.

Meng ZENG, Na ZHAO, Xianqi WANG, et al. Establishment of a chest-CT-image-based classification deep learning model for pneumonia prediction[J]. Journal of Army Medical University, 2023, 45(21): 2266-2274. DOI： 10.16016/j.2097-0927.202307139.

摘要

目的

建立基于胸部CT图像的AI模型

实现对细菌、真菌、病毒性(包括COVID-19)肺炎的快速分类预测。

方法

回顾性收集2013-2020年就诊于陆军军医大学附属第一医院的559例细菌性、真菌性及非COVID-19病毒性肺炎患者及2020年就诊于重庆大学附属三峡医院的53例COVID-19患者的影像资料

首先利用Resnet_18、Efficientnet_b5、ViT、Swin-Transformer等4种典型的深度神经网络构建图像级三分类及四分类预测模型

在独立测试集中进行验证选出最优模型; 然后分析采用单张图像和3张图像融合构建数据集对模型的影响; 最后分别使用按图像类别占比投票及随机森林2种方法进行患者级分类预测。使用精确率、召回率、特异性、F1值、AUC、准确率评估模型效能

最终筛选出表现最优的AI预测模型。

结果

在图像级分类中Swin-Transformer模型表现最佳

三分类准确率为0.932

四分类准确率为0.948。测试Swin-Transformer模型分别采用单张图像和3张图像融合构建数据集的效果

采用融合图像的Swin-Transformer_C模型效能进一步提升

在测试集中三分类准确率和AUC值分别为0.931、0.989

四分类准确率和AUC值分别为0.952、0.990。使用Swin-Transformer_C模型进行患者级分类

采用随机森林的方法预测效能更佳

三分类准确率和AUC值分别为0.984、0.987

四分类准确率和AUC值分别为0.967、0.971。Resnet_18、Efficientent_b5、Vit等3种网络也取得了较好的效果

但总体效能低于Swin-Transformer网络。

结论

基于融合数据建立的深度学习模型Swin-Transformer_C与其他4种模型相比在图像级分类中效果最佳

其融合随机森林分类器在患者级分类中也取得最优性能。表明深度学习可用于不同病原体感染所致肺炎类型的快速分类预测。

Abstract

Objective

To establish an AI model based on chest CT images to achieve rapid classification prediction for bacterial

fungal

and viral (including COVID-19) infectious pneumonia.

Methods

Chest CT imaging data of 559 bacterial

fungal and non-COVID-19 viral pneumonia patients admitted to the First Affiliated Hospital of Army Medical University from 2013 to 2020 and 53 COVID-19 patients to the Chongqing University Three Gorges Hospital from January 2020 to December 2020 were collected and analyzed retrospectively.Firstly

4 typical deep neural networks (Resnet_18

Efficientnet_b5

ViT

and Swin-Transformer) were used to construct image-level triple and quadruple classification prediction models

and the optimal model was selected by validation in an independent test set.Then

the effects of single-image and three-fused-images to construct dataset on the models were analyzed.Finally

voting by image category and random forest were carried out to make the patient-level classification prediction

respectively.The precision

recall rate

specificity

F1 value

AUC

and accuracy were employed to evaluate the performance of the models in order to screen out the best performing AI prediction model.

Results

The Swin Transformer model performed best in image-level classification

with a triple classification accuracy of 0.932 and a quadruple classification accuracy of 0.948.After the model was constructed with single-image and three-fused-images

a model

named as Swin-Transformer_C

which was further improved with fused images

showed good performance

with a triple classification accuracy and AUC of 0.931 and 0.989

and with a quadruple classification accuracy and AUC of 0.952 and 0.990

respectively

in the test set.Patient-level categorization using the Swin-transformer_C model integrated with random forest was more effective

with a triple classification accuracy and AUC of 0.984 and 0.987

and with a quadruple classification accuracy and AUC of 0.967 and 0.971

respectively.Three other networks

Resnet_18

Efficientent_b5 and Vit

also achieved good results

but the overall effectiveness was lower than that of the Swin-transformer network.

Conclusion

The deep learning model Swin-transformer_C

built on fused dataset

shows best performance in image-level classification when compared to the other 4 models

and it achieves optimal performance by integrating with random forest classifier in patient-level classification.Our study demonstrates that deep learning can be used for rapid classification prediction of pneumonia types due to different pathogen infections.

关键词

Keywords

references

MURDOCH D R, HOWIE S R C. The global burden of lower respiratory infections: making progress, but we need to do better[J]. Lancet Infect Dis, 2018, 18(11): 1162-1163. DOI: 10.1016/S1473-3099(18)30407-9.

ROTH G A, ABATE D, ABATE K H, et al . Global, regional, and national age-sex-specific mortality for 282 causes of death in 195 countries and territories, 1980-2017: a systematic analysis for the Global Burden of Disease Study 2017[J ] . Lancet, 2018, 392(10159): 1736-1788. DOI: 10.1016/S0140-6736(18)32203-7.

PERUMAL V, NARAYANAN V, RAJASEKAR S J S. Detection of COVID-19 using CXR and CT images using Transfer Learning and Haralick features[J]. Appl Intell, 2021, 51(1): 341-358. DOI: 10.1007/s10489-020-01831-z.

LONG C Q, XU H X, SHEN Q L, et al . Diagnosis of the coronavirus disease (COVID-19): rRT-PCR or CT?[J ] . Eur J Radiol, 2020, 126: 108961. DOI: 10.1016/j.ejrad.2020.108961.

LEE E Y P, NG M Y, KHONG P L. COVID-19 pneumonia: what has CT taught us?[J]. Lancet Infect Dis, 2020, 20(4): 384-385. DOI: 10.1016/S1473-3099(20)30134-1.

LI L, QIN L X, XU Z G, et al . Using artificial intelligence to detect COVID-19 and community-acquired pneumonia based on pulmonary CT: evaluation of the diagnostic accuracy[J ] . Radiology, 2020, 296(2): E65-E71. DOI: 10.1148/radiol.2020200905.

ZHANG K, LIU X H, SHEN J, et al . Clinically applicable AI system for accurate diagnosis, quantitative measurements, and prognosis of COVID-19 pneumonia using computed tomography[J ] . Cell, 2020, 181(6): 1423-1433. e11. DOI: 10.1016/j.cell.2020.04.045.

KERMANY D S, GOLDBAUM M, CAI W J, et al . Identifying medical diagnoses and treatable diseases by image-based deep learning[J ] . Cell, 2018, 172(5): 1122-1131. e9. DOI: 10.1016/j.cell.2018.02.010.

RAJARAMAN S, CANDEMIR S, KIM I, et al . Visualization and interpretation of convolutional neural network predictions in detecting pneumonia in pediatric chest radiographs[J ] . Appl Sci(Basel), 2018, 8(10): 1715. DOI: 10.3390/app8101715.

DU W, LUO X J, CHEN M. A practical deep learning model in differentiating pneumonia-type lung carcinoma from pneumonia on CT images: ResNet added with attention mechanism[J]. J Oncol, 2022, 2022: 1-9. DOI: 10.1155/2022/8906259.

IBRAHIM A U, OZSOZ M, SERTE S, et al . Pneumonia classification using deep learning from chest X-ray images during COVID-19[J ] . Cogn Comput, 2021: 1-13. DOI: 10.1007/s12559-020-09787-5.

ABBAS A, ABDELSAMEA M M, GABER M M. Classification of COVID-19 in chest X-ray images using DeTraC deep convolutional neural network[J]. Appl Intell(Dordr), 2021, 51(2): 854-864. DOI: 10.1007/s10489-020-01829-7.

BARSHOOI A H, AMIRKHANI A. A novel data augmentation based on Gabor filter and convolutional deep learning for improving the classification of COVID-19 chest X-Ray images[J]. Biomed Signal Process Control, 2022, 72: 103326. DOI: 10.1016/j.bspc.2021.103326.

HE K M, ZHANG X Y, REN S Q, et al . Deep residual learning for image recognition[C ] //2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). June 27-30, 2016. Las Vegas: IEEE, 2016: 770-778. DOI: 10.1109/CVPR.2016.90.

WU N, PHANG J, PARK J, et al . Deep neural networks improve radiologists' performance in breast cancer screening[J ] . IEEE Trans Med Imaging, 2020, 39(4): 1184-1194. DOI: 10.1109/TMI.2019.2945514.

KARTHIK R, MENAKA R, HARIHARAN M. Learning distinctive filters for COVID-19 detection from chest X-ray using shuffled residual CNN[J]. Appl Soft Comput, 2021, 99: 106744. DOI: 10.1016/j.asoc.2020.106744.

TAN M, LE Q. Efficientnet: Rethinking model scaling for convolutional neural networks[EB/OL]. [2020-09-11]. http://doi.org/10.48550/arXiv.1905.11946.

MARQUES G, AGARWAL D, DE LA TORRE DÍEZ I. Automated medical diagnosis of COVID-19 through EfficientNet convolutional neural network[J]. Appl Soft Comput, 2020, 96: 106691. DOI: 10.1016/j.asoc.2020.106691.

WANG J, LIU Q Y, XIE H T, et al . Boosted EfficientNet: detection of lymph node metastases in breast cancer using convolutional neural networks[J ] . Cancers, 2021, 13(4): 661. DOI: 10.3390/cancers13040661.

DOSOVITSKIY A, BEYER L, KOLESNIKOV A, et al . An image is worth 16x16 words: transformers for image recognition at scale[EB/OL ] . [2021-06-03 ] . https://doi.org/10.485501/arXiv.2010.11929.

CHEN B, LIU Y, ZHANG Z, et al . Transattunet: Multi-level attention-guided u-net with transformer for medical image segmentation[EB/OL ] . [2022-07-09 ] . https://doi.org/10.48550/arXiv.2017.05274.

LIU Z, LIN Y T, CAO Y, et al . Swin transformer: hierarchical vision transformer using shifted windows[C ] //2021 IEEE/CVF International Conference on Computer Vision (ICCV). [2021-10-17 ] . Montreal: IEEE, 2022: 9992-10002. DOI: 10.1109/ICCV48922.2021.00986.

LEE H H, BAO S X, HUO Y K, et al . 3D UX-net: a large kernel volumetric ConvNet modernizing hierarchical transformer for medical image segmentation[EB/OL ] . [2023-05-02 ] . https:// doi.org/10.48550./arXiv.2209.15076.

DANGIS A, GIERAERTS C, DE BRUECKER Y, et al . Accuracy and reproducibility of low-dose submillisievert chest CT for the diagnosis of COVID-19[J ] . Radiol Cardiothorac Imaging, 2020, 2(2): e200196. DOI: 10.1148/ryct.2020200196.

MAJIDI H, NIKSOLAT F. Chest CT in patients suspected of COVID-19 infection: a reliable alternative for RT-PCR[J]. Am J Emerg Med, 2020, 38(12): 2730-2732. DOI: 10.1016/j.ajem.2020.04.016.

LIU C L, WANG X Y, LIU C B, et al . Differentiating novel coronavirus pneumonia from general pneumonia based on machine learning[J ] . Biomed Eng Online, 2020, 19: 66. DOI: 10.1186/s12938-020-00809-9.

周奇浩, 张俊华, 普钟, 等. 结合深层密集聚合的新冠肺炎CT图像分类方法[J]. 计算机应用研究, 2023, 40(6): 1857-1863. DOI: 10.19734/j.issn.1001-3695.2022.08.0502.

ZHOU Q H, ZHANG J H, PU Z, et al . COVID-19 CT image classification method combined with deep layer dense aggregation[J ] . Appl Res Comput, 2023, 40(6): 1857-1863. DOI: 10.19734/j.issn.1001-3695.2022.08.0502.

KASSANIA S H, KASSANIB P H, WESOLOWSKIC M J, et al . Automatic detection of coronavirus disease (COVID-19) in X-ray and CT images: a machine learning based approach[J ] . Biocybern Biomed Eng, 2021, 41(3): 867-879. DOI: 10.1016/j.bbe.2021.05.013.

张淙越, 杨晓玲. 基于卷积神经网络的新冠肺炎CT图像识别系统[J]. 电脑与信息技术, 2022, 30(3): 12-14, 40. DOI: 10.19414/j.cnki.1005-1228.2022.03.013.

ZHANG C Y, YANG X L. COVID-19 CT image recognition system based on convolutional neural network[J]. Comput Inf Technol, 2022, 30(3): 12-14, 40. DOI: 10.19414/j.cnki.1005-1228.2022.03.013.

WANG G Y, LIU X H, SHEN J, et al . A deep-learning pipeline for the diagnosis and discrimination of viral, non-viral and COVID-19 pneumonia from chest X-ray images[J ] . Nat Biomed Eng, 2021, 5(6): 509-521. DOI: 10.1038/s41551-021-00704-1.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

人工智能在临床领域的研究进展及前景展望

基于贝叶斯网络算法建立全膝关节置换术后认知功能障碍风险预测模型

基于两样本孟德尔随机化与可解释机器学习算法的冠心病风险预测模型构建及发病因素探究

融合深度学习、临床与影像组学特征的综合模型提升非小细胞肺癌EGFR突变预测效能

基于系统评价和Meta分析的糖尿病足溃疡患者截肢风险预测模型的开发和验证