KIPS Transactions on Computer and Communication Systems (정보처리학회논문지:컴퓨터 및 통신 시스템)

Korea Information Processing Society (한국정보처리학회)
권호 표지
DOI QR코드

DOI QR Code

Development of Cloud-Based Medical Image Labeling System and It's Quantitative Analysis of Sarcopenia

클라우드기반 의료영상 라벨링 시스템 개발 및 근감소증 정량 분석

  • 이충섭 (원광대학교 의료융합연구센터) ;
  • 임동욱 (원광대학교 의료융합연구센터) ;
  • 김지언 (원광대학교 의료융합연구센터) ;
  • 노시형 (원광대학교 의료융합연구센터) ;
  • 유영주 (세종사이버대학교 소프트웨어공학과) ;
  • 김태훈 (원광대학교병원 스마트사업팀) ;
  • 윤권하 (성균관대학교 삼성창원병원 영상의학과) ;
  • 정창원 (원광대학교병원 스마트사업팀)
  • Received : 2021.12.30
  • Accepted : 2022.03.29
  • Published : 2022.07.31
Download PDF
⟨ Previous Next ⟩

Abstract

Most of the recent AI researches has focused on developing AI models. However, recently, artificial intelligence research has gradually changed from model-centric to data-centric, and the importance of learning data is getting a lot of attention based on this trend. However, it takes a lot of time and effort because the preparation of learning data takes up a significant part of the entire process, and the generation of labeling data also differs depending on the purpose of development. Therefore, it is need to develop a tool with various labeling functions to solve the existing unmetneeds. In this paper, we describe a labeling system for creating precise and fast labeling data of medical images. To implement this, a semi-automatic method using Back Projection, Grabcut techniques and an automatic method predicted through a machine learning model were implemented. We not only showed the advantage of running time for the generation of labeling data of the proposed system, but also showed superiority through comparative evaluation of accuracy. In addition, by analyzing the image data set of about 1,000 patients, meaningful diagnostic indexes were presented for men and women in the diagnosis of sarcopenia.

최근 대부분의 인공지능 연구는 AI 모델 개발에 중점을 두고 있다. 하지만 최근 인공지능 연구가 모델 중심에서 데이터 중심으로 점차 변경되고 이런 추세를 바탕으로 학습데이터의 중요성이 크게 주목 받고 있다. 그러나 학습데이터의 준비과정이 전체 과정의 상당 부분을 차지하고 라벨링 데이터 생성 또한 개발 목적에 따라 다르기 때문에 많은 시간과 노력이 필요하다. 따라서 기존의 미충족을 해결하기 위한 다양한 라벨링 기능을 갖는 도구 개발이 필요하다. 본 논문에서는 의료영상의 라벨링 데이터를 정교하고 빠르게 생성하기 위한 라벨링 시스템에 대해서 기술한다. 이를 구현하기 위해서 Back Projection, GrabCut 기법을 이용한 반자동 방식과 기계학습 모델을 통해서 예측한 자동 방식의 라벨링 기능을 구현하였다. 우리는 제안한 시스템의 라벨링 데이터 생성에 대한 수행시간의 장점을 보였을뿐만 아니라 정확성에 대한 비교평가를 통해 우수성을 보였다. 또한 1,000여명의 환자 영상 데이터셋을 분석하여 근감소증 진단에 남성과 여성에 의미있는 진단지표를 제시하였다.

Keywords

Acknowledgement

본 연구는 보건복지부의 재원으로 한국보건산업진흥원의 보건의료기술 연구개발사업 (HI18C1216), 그리고 한국연구재단(NRF-2021R1A5A8029876) (NRF-2020R1I1A1A01074256)의 지원으로 수행함.

References

  1. Andrew Ng: MLOps: "From Model-centric to Data-centric AI," [Internet], https://www.deeplearning.ai/wp-content/uploads/2021/06/MLOps-From-Model-centric-to-Data-centric-AI.pdf.
  2. Y. Bai, et al., "How important is the train-validation split in meta-learning?" arXiv preprint arXiv:2010.05843, 2020.
  3. A. Racz, D. Bajusz, and K. Heberger, "Effect of dataset size and Train/Test split ratios in QSAR/QSPR multiclass classification," Molecules, Vol.26, No.4, pp.1111, 2021. https://doi.org/10.3390/molecules26041111
  4. M. Motamedi, N. Sakharnykh, and T. Kaldewey, "A datacentric approach for training deep neural networks with less data," arXiv preprint arXiv:2110.03613, 2021.
  5. J. D. Bosser, E. Sorstadius, and M. H. Chehreghani, "Modelcentric and data-centric aspects of active learning for neural network models," arXiv preprint arXiv:2009.10835, 2020.
  6. Cognilytica, "Data engineering, preparation, and labeling for AI 2019," [Internet], https://www.cognilytica.com/2019/03/06/report-data-engineering-preparation-and-abeling-for-ai-2019/.
  7. Y. Roh, G. Heo, and S. E. Whang, "A survey on data collection for machine learning: A big data-ai integration perspective," in IEEE Transactions on Knowledge and Data Engineering, Vol.33, No.4, pp.1328-1347, 2021, doi: 10.1109/TKDE.2019.2946162.
  8. Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," Nature, Vol.521, pp.436-444, 2015. https://doi.org/10.1038/nature14539
  9. Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, "Gradient-based learning applied to document recognition," Proceedings of the IEEE, Vol.86, No.11, pp.2278-2324, 1998. https://doi.org/10.1109/5.726791
  10. A. Krizhevsky, "Learning multiple layers of features from tiny images," [Internet], https://www.cs.toronto.edu/~kriz/cifar.html, 2009.
  11. M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman, "The PASCAL Visual Object Classes (VOC) Challenge," International Journal of Computer Vision, Vol.88, pp.303-338, 2010. https://doi.org/10.1007/s11263-009-0275-4
  12. T. Y. Lin, et al., "Microsoft COCO: Common objects in context," In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds) Computer Vision ? ECCV 2014. Lecture Notes in Computer Science, Vol.8693. Springer, Cham, doi: 10.1007/978-3-319-10602-1_48
  13. J. Deng, W. Dong, R. Socher, L. J. Li, K. Li, and L. Fei-Fe, "Imagenet: A large-scale hierarchical image database," 2009 IEEE Conference on Computer Vision and Pattern Recognition, pp.248-255, 2009.
  14. Google's Open Images V6 + Extentions [Internet], https://storage.googleapis.com/openimages/web/index.html.
  15. Center for Artificial Intelligence in Medicine & Imaging [Internet], https://aimi.stanford.edu/.
  16. AI Hub [Internet], https://aihub.or.kr/.
  17. Pathology AI Platform, [Internet], http://wisepaip.org/.
  18. Morphometry Open AI Innovation [Internet], http://www.wonmoai.org/.
  19. A. D. Weston, et al., "Automated abdominal segmentation of CT scans for body composition analysis using deep learning," Radiology, Vol.290, No.3, pp.669-679, 2019. doi: 10.1148/radiol.2018181432.
  20. C. T. Rueden, et al., "ImageJ2: ImageJ for the next generation of scientific image data," BMC Bioinformatics, Vol.18, No.529, 2017. doi: 10.1186/s12859-017-1934-z.
  21. R. Kikinis, S. D. Pieper, and K. G. Vosburgh, "3D Slicer: A platform for subject-specific image analysis, visualization, and clinical support," In: Jolesz, F. (eds.) Intraoperative Imaging and Image-Guided Therapy. Springer, New York, NY. doi: 10.1007/978-1-4614-7657-3_19.
  22. M. N. Rizve, K., Duarte, Y. S., Rawat, and M. Shah, "In defense of pseudo-labeling: An uncertainty-aware pseudo-label selection framework for semi-supervised learning," arXiv preprint arXiv:2101.06329, 2021.
  23. S. Kim, T. H. Kim, C. W. Jeong, C. Lee, S. Noh, J. E. Kim, and K. H. Yoon, "Development of quantification software for evaluating body composition contents and its clinical application in sarcopenic obesity," Scientific Reports, Vol.10, No.10452, 2020. doi: 10.1038/s41598- 020-67461-0.
  24. Histogram Backprojection [Internet], https://docs.opencv.org/3.4/dc/df6/tutorial_py_histogram_backprojection.html.
  25. GrabCut [Internet], https://docs.opencv.org/3.4/d8/d83/tutorial_py_grabcut.html.
  26. E. H. Kim, et al., "Reference data and t-scores of lumbar skeletal muscle area and its skeletal muscle indices measured by CT scan in a healthy Korean population," The Journals of Gerontology: Series A, Vol.76, No.2, pp.265-271, 2021. doi: 10.1093/gerona/glaa065.