한국컴퓨터그래픽스학회논문지 (Journal of the Korea Computer Graphics Society)

  • 제30권4호
  • /
  • Pages.19-29
  • /
  • 2024
  • /
  • 1975-7883(pISSN)
  • /
  • 2383-529X(eISSN)
한국컴퓨터그래픽스학회 (Korea Computer Graphics Society)
권호 표지
DOI QR코드

DOI QR Code

의료영상 분류를 위한 심층신경망 훈련에서 StyleGAN 합성 영상의 데이터 증강 효과 분석

Data Augmentation Effect of StyleGAN-Generated Images in Deep Neural Network Training for Medical Image Classification

  • 이한상 (한국과학기술원 정보전자연구소) ;
  • 우아라 (서울여자대학교 소프트웨어융합학과) ;
  • 홍헬렌 (서울여자대학교 소프트웨어융합학과)
  • Hansang Lee (School of Electrical Engineering, Information & Electronics Research Institute, KAIST) ;
  • Arha Woo (Department of Software Convergence, Seoul Women's University) ;
  • Helen Hong (Department of Software Convergence, Seoul Women's University)
  • 투고 : 2024.06.08
  • 심사 : 2024.08.13
  • 발행 : 2024.09.01
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 논문에서는 의료 영상 분류를 위한 심층 신경망 훈련에서 StyleGAN 합성 영상의 데이터 증강 효과를 분석한다. 이를 위해 흉부 X선 영상에서의 폐렴 진단과 복부 CT 영상에서의 간전이암 분류 문제에서 StyleGAN 합성 영상을 이용하여 VGG-16 심층 합성곱 신경망 훈련을 수행한다. 실험에서 분류 결과에 대한 정량적, 정성적 분석을 통해 StyleGAN 데이터 증강이 특징 공간에서 클래스 외곽을 확장하는 특성을 보이며, 이와 같은 특성으로 인해 실제 영상과의 적절한 비율을 통해 혼합했을 때 분류 성능이 개선될 수 있음을 확인하였다.

In this paper, we examine the effectiveness of StyleGAN-generated images for data augmentation in training deep neural networks for medical image classification. We apply StyleGAN data augmentation to train VGG-16 networks for pneumonia diagnosis from chest X-ray images and focal liver lesion classification from abdominal CT images. Through quantitative and qualitative analyses, our experiments reveal that StyleGAN data augmentation expands the outer class boundaries in the feature space. Thanks to this expansion characteristics, the StyleGAN data augmentation can enhance classification performance when properly combined with real training images.

키워드

과제정보

이 연구는 정부(과학기술정보통신부)의 재원으로 한국연구재단의 지원을 받아 수행된 연구(No. RS-2023-00207947)이며, 서울여자대학교 학술연구비의 지원에 의한 것임(2024-0220). 본 논문에서 사용한 복부 CT 영상 데이터를 제공해주신 세브란스병원 영상의학과 임준석 교수님께 감사의 말씀을 드립니다.

참고문헌

  1. J. Garstka and M. Strzelecki, "Pneumonia detection in x-ray chest images based on convolutional neural networks and data augmentation methods," in 2020 Signal Processing: Algorithms, Architectures, Arrangements, and Applications. Institute of Electrical and Electronics Engineers, 2020, pp. 18-23.
  2. S. Motamed, P. Rogalla, and F. Khalvati, "Data augmentation using generative adversarial networks (gans) for gan-based detection of pneumonia and covid-19 in chest x-ray images," Informatics in Medicine Unlocked, vol. 27, p. 100779, 2021.
  3. A. Radford, L. Metz, and S. Chintala, "Unsupervised representation learning with deep convolutional generative adversarial networks," in International Conference on Learning Representations, 2016.
  4. T. Karras, T. Aila, S. Laine, and J. Lehtinen, "Progressive growing of gans for improved quality, stability, and variation," in International Conference on Learning Representations, 2017.
  5. T. Karras, S. Laine, and T. Aila, "A style-based generator architecture for generative adversarial networks," in Computer Vision and Pattern Recognition, 2019.
  6. H. Salehinejad, S. Valaee, T. Dowdell, E. Colak, and J. Barfett, "Generalization of deep neural networks for chest pathology classification in x-rays using generative adversarial networks," in Institute of Electrical and Electronics Engineer international conference on acoustics, speech and signal processing, 2018.
  7. 김덕선, 우아라, 이한상, and 홍헬렌, "복부 ct 영상에서 심층 합성곱 신경망 기반의 국소 간 병변 분류를 위한 데이터 증강 기법의 효과," 한국컴퓨터그래픽스학회논문지, vol. 29, no. 2, pp. 1-11, 2023.
  8. D. Zhao, D. Zhu, J. Lu, Y. Luo, and G. Zhang, "Synthetic medical images using f&bgan for improved lung nodules classification by multi-scale vgg-16," Symmetry, vol. 10, no. 10, p. 519, 2018.
  9. H. Lee, H. Lee, H. Hong, H. Bae, J. Lim, and J. Kim, "Classification of focal liver lesions in ct images using convolutional neural networks with lesion information augmented patches and synthetic data augmentation," Medical Physics, vol. 48, no. 9, pp. 5029-5046, 2021.
  10. A. Woo, H. Lee, J. Lim, and H. Hong, "Classification of focal liver lesions in abdominal ct images using convolutional neural networks with stylegan data augmentation," in Proceeding of the Fall Conference of the Korea Multimedia Society, vol. 26, no. 1, 2023, pp. 94-96.
  11. M. Loey, S. Florentin, and N. Khalifa, "Within the lack of chest covid-19 x-ray dataset: a novel detection model based on gan and deep transfer learning," Symmetry, vol. 12, no. 4, p. 651, 2020.
  12. S. Buragadda, K. Rani, S. Vasantha, and M. Chakravarthi, "Hcugan: Hybrid cyclic unet gan for generating augmented synthetic images of chest x-ray images for multi classification of lung diseases," International Journal of Engineering Trends and Technology, vol. 70, no. 2, pp. 229-238, 2022.
  13. L. van der Maaten and G. Hinton, "Visualizing data using t-sne," Journal of Machine Learning Research, vol. 9, pp. 2579-2605, 2008.
  14. R. Hao, K. Namdar, L. Liu, M. Haider, and F. Khalvati, "A comprehensive study of data augmentation strategies for prostate cancer detection in diffusion-weighted mri using convolutional neural networks," Journal of Digital Imaging, pp. 862-876, 2021.
  15. M. Kim and H. Bae, "Data augmentation techniques for deep learning based medical image analyses," Journal of the Korean Society of Radiology, vol. 81, no. 6, 2020.
  16. M. Nishio, S. Noguchi, H. Matsuo, and T. Murakami, "Automatic classification between covid-19 pneumonia, non-covid19 pneumonia, and the healthy on chest x-ray image: combination of data augmentation methods," Scientific Reports, vol. 10, no. 1, pp. 1-6, 2020.
  17. D. Kermany, M. Goldbaum, W. Cai, C. Valentim, H. Liang, S. Baxter, et al., "Identifying medical diagnoses and treatable diseases by image-based deep learning," Cell, vol. 172, no. 5, pp. 1122-1131, 2018.
  18. K. Asnaoui, Y. Chawki, and A. Idri, "Automated methods for detection and classification pneumonia based on x-ray images using deep learning," in Artificial intelligence and blockchain for future cybersecurity applications. Cham: Springer International Publishing, 2021, pp. 257-284.
  19. I. Rudan, C. Boschi-Pinto, Z. Biloglav, K. Mulholland, and H. Campbell, "Epidemiology and etiology of childhood pneumonia," Bulletin of the World Health Organization, vol. 86, pp. 408-416B, 2008.
  20. H. Zar, S. Andronikou, and P. Nicol, "Advances in the diagnosis of pneumonia in children," British Medical Journal, vol. 358, 2017.