Journal of radiological science and technology (대한방사선기술학회지:방사선기술과학)

Korean Society of Radiological Science (대한방사선과학회)
권호 표지
DOI QR코드

DOI QR Code

Performance Evaluation of ResNet-based Pneumonia Detection Model with the Small Number of Layers Using Chest X-ray Images

흉부 X선 영상을 이용한 작은 층수 ResNet 기반 폐렴 진단 모델의 성능 평가

  • Youngeun Choi (Department of Medical Science, Konyang University) ;
  • Seungwan Lee (Department of Medical Science, Konyang University)
  • Received : 2023.06.14
  • Accepted : 2023.08.03
  • Published : 2023.08.31
Download PDF
⟨ Previous Next ⟩

Abstract

In this study, pneumonia identification networks with the small number of layers were constructed by using chest X-ray images. The networks had similar trainable-parameters, and the performance of the trained models was quantitatively evaluated with the modification of the network architectures. A total of 6 networks were constructed: convolutional neural network (CNN), VGGNet, GoogleNet, residual network with identity blocks, ResNet with bottleneck blocks and ResNet with identity and bottleneck blocks. Trainable parameters for the 6 networks were set in a range of 273,921-294,817 by adjusting the output channels of convolution layers. The network training was implemented with binary cross entropy (BCE) loss function, sigmoid activation function, adaptive moment estimation (Adam) optimizer and 100 epochs. The performance of the trained models was evaluated in terms of training time, accuracy, precision, recall, specificity and F1-score. The results showed that the trained models with the small number of layers precisely detect pneumonia from chest X-ray images. In particular, the overall quantitative performance of the trained models based on the ResNets was above 0.9, and the performance levels were similar or superior to those based on the CNN, VGGNet and GoogleNet. Also, the residual blocks affected the performance of the trained models based on the ResNets. Therefore, in this study, we demonstrated that the object detection networks with the small number of layers are suitable for detecting pneumonia using chest X-ray images. And, the trained models based on the ResNets can be optimized by applying appropriate residual-blocks.

Keywords

References

  1. Quekel LG, Kessels AG, Goei R, Van Engelshoven MA. Miss rate of lung cancer on the chest radiograph in clinical practice. Chest. 1999;15(3):720-4. https://doi.org/10.1378/chest.115.3.720
  2. Hong JY, Park SH, Jung YJ. Artificial Intelligence based medical imaging: An overview. Journal of Radiological Science and Technology. 2020;43(3):195-208. https://doi.org/10.17946/JRST.2020.43.3.195
  3. Kim YJ, Kim KG. Development of an optimized deep learning model for medical imaging. Journal of the Korean Society of Radiology. 2020;81(6):1274-89. https://doi.org/10.3348/jksr.2020.0171
  4. Oh JY, Jeong EH, Lee JY, Park HH. Evaluating usefulness of deep learning based left ventricle segmentation in cardiac gated blood pool scan. Journal of Radiological Science and Technology. 2022;45(2):151-8. https://doi.org/10.17946/JRST.2022.45.2.151
  5. Seah JCY, Tang CHM, Buchlak QD, Holt XG, Wardman JB, Aimoldin A, et al. Effect of a comprehensive deep-learning model on the accuracy of chest x-ray interpretation by radiologists: A retrospective, multireader multicase study. The Lancet Digital Health. 2021;3(8):496-506. https://doi.org/10.1016/S2589-7500(21)00106-0
  6. Zakaria R, Abdelmajid H, Zitouni D. Deep learning in medical imaging: A review. Applications of Machine Intelligence in Engineering. 2022:131-44.
  7. Anis S, Lai KW, Chuah JH, Ali SM, Mohafez H, Hadizadeh M, et al. An overview of deep learning approaches in chest radiograph. IEEE Access. 2020;8:182347-54. https://doi.org/10.1109/ACCESS.2020.3028390
  8. Aslani S, Jacob J. Utilisation of deep learning for COVID-19 diagnosis. Clinical Radiology. 2023;78(2):150-7. https://doi.org/10.1016/j.crad.2022.11.006
  9. Hussain E, Hasan M, Rahman MA, Lee I, Tamanna T, Parvez MZ. CoroDet: A deep learning based classification for COVID-19 detection using chest X-ray images. Chaos, Solitons & Fractals. 2021;142:110495.
  10. Ikechukwu AV, Murali S, Deepu R, Shivamurthy RC. ResNet-50 vs VGG-19 vs training from scratch: A comparative analysis of the segmentation and classification of Pneumonia from chest X-ray images. Global Transitions Proceedings. 2021;2(2):375-81. https://doi.org/10.1016/j.gltp.2021.08.027
  11. He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016:770-8.
  12. Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition. arXiv: 1409.1556; 2014.
  13. Ying X. An overview of overfitting and its solutions. Journal of Physics: Conference Series. 2019;1168(2):022022.
  14. Yuan P, Lin S, Cui C, Du Y, Guo R, He D, et al. HS-ResNet: Hierarchical-split block on convolutional neural network. arXiv:2010.07621; 2020.
  15. Amjoud AB, Amrouch M. Convolutional neural networks backbones for object detection. Proceedings of the International Conference on Image and Signal Processing. 2020:282-9.
  16. He T, Zhang Z, Zhang H, Zhang Z, Xie J, Li M. Bag of tricks for image classification with convolutional neural networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2019:558-67.
  17. Taylor L, Nitschke G. Improving deep learning with generic data augmentation. IEEE Symposium Series on Computational Intelligence. 2018:1542-7.
  18. Kaggle. Pneumonia classification. [cited 2023 May 12]. Available from https://www.kaggle.com/datasets/paultimothymooney/chest-xray-pneumonia.
  19. Shorten C, Khoshgoftaar TM. A survey on image data augmentation for deep learning. Journal of Big Data. 2019;6(1): 1-48. https://doi.org/10.1186/s40537-019-0197-0
  20. Zagoruyko S, Komodakis N. Wide residual networks. arXiv: 1605.0714; 2016.
  21. Gu J, Wang Z, Kuen J, Ma L, Shahroudy A, Shuai B, et al. Recent advances in convolutional neural networks. Pattern Recognition. 2018;77:354-77. https://doi.org/10.1016/j.patcog.2017.10.013
  22. Szegedy C, Liu W, Jia Y, Sermanet P, Reed S, Anguelov D, et al. Going deeper with convolutions. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2015:1-9.
  23. Loshchilov I, Hutter F. Decoupled weight decay regularization. arXiv: 1711.05101; 2017.
  24. Dhumale RB, Thombare ND, Bangare PM. Machine learning: A way of dealing with Artificial Intelligence. International Conference on Innovations in Information and Communication Technology. 2019:1-6.
  25. Gholamalinezhad H, Khosravi H. Pooling methods in deep neural networks, a review. arXiv: 2009.07485; 2020.
  26. Jiao Y, Du P. Performance measures in evaluating machine learning based bioinformatics predictors for classifications. Quantitative Biology. 2016;4:320-30. https://doi.org/10.1007/s40484-016-0081-2
  27. Lin TY, Goyal P, Girshick R, He K, Dollar P. Focal loss for dense object detection. Proceedings of the IEEE International Conference on Computer Vision. 2017:2980-8.
  28. Nagi J, Ducatelle F, Di Caro GA, Ciresan D, Meier U, Giusti A, et al. Max-pooling convolutional neural networks for vision-based hand gesture recognition. IEEE International Conference on Signal and Image Processing Applications. 2011:342-7.
  29. Guler O, Polat K. Classification performance of deep transfer learning methods for pneumonia detection from chest X-ray images. Journal of Artificial Intelligence and Systems. 2022;4(1):107-26.
  30. Chouhan V, Singh SK, Khamparia A, Gupta D, Tiwari P, Moreira C, et al. A novel transfer learning based approach for pneumonia detection in chest X-ray images. Applied Sciences. 2020;10(2):559.
  31. Ali F, Khan S, Abbas AW, Shah B, Hussain T, Song D, et al. A two-tier framework based on GoogLeNet and YOLOv3 Models for tumor detection in MRI. Computers, Materials and Continua. 2022;72:73-92. https://doi.org/10.32604/cmc.2022.024103
  32. Han Y, Chen C, Tewfik A, Ding Y, Peng Y. Pneumonia detection on chest x-ray using radiomic features and contrastive learning. IEEE International Symposium on Biomedical Imaging. 2021:247-51.
  33. Yu Z, Dong Y, Cheng J, Sun M, Su F. Research on face recognition classification based on improved GoogleNet. Security and Communication Networks. 2022;2022:1-6. https://doi.org/10.1155/2022/7192306