DOI QR코드

DOI QR Code

Urinary Stones Segmentation Model and AI Web Application Development in Abdominal CT Images Through Machine Learning

기계학습을 통한 복부 CT영상에서 요로결석 분할 모델 및 AI 웹 애플리케이션 개발

  • Received : 2021.07.06
  • Accepted : 2021.08.31
  • Published : 2021.11.30

Abstract

Artificial intelligence technology in the medical field initially focused on analysis and algorithm development, but it is gradually changing to web application development for service as a product. This paper describes a Urinary Stone segmentation model in abdominal CT images and an artificial intelligence web application based on it. To implement this, a model was developed using U-Net, a fully-convolutional network-based model of the end-to-end method proposed for the purpose of image segmentation in the medical imaging field. And for web service development, it was developed based on AWS cloud using a Python-based micro web framework called Flask. Finally, the result predicted by the urolithiasis segmentation model by model serving is shown as the result of performing the AI web application service. We expect that our proposed AI web application service will be utilized for screening test.

의료분야 인공지능 기술이 분석과 알고리즘 개발에 중점을 두었으나 점차 제품으로 서비스하기 위한 Web 애플리케이션 개발로 변화되고 있다. 본 연구는 복부 CT 영상에서 요로결석(Urinary Stone) 분할모델과 이를 기반으로 한 인공지능 웹 애플리케이션에 대해 기술한다. 이를 구현하기 위해 의료영상 분야에서 이미지 분할을 목적으로 제안된 End-to-End 방식의 Fully-Convolutional Network 기반 모델인 U-Net을 사용하여 모델을 개발하였다. 그리고 Python 기반의 Flask라는 마이크로 웹 프레임워크를 사용하여 AWS 클라우드 기반 웹 애플리케이션으로 개발하였다. 끝으로 모델 서빙으로 요로결석 분할모델이 예측한 결과를 인공지능 웹 애플리케이션 서비스 수행 결과로 보인다. 제안한 AI 웹 애플리케이션 서비스가 선별 검사에 활용되기를 기대한다.

Keywords

Acknowledgement

본 연구는 보건복지부의 재원으로 한국보건산업진흥원의 보건의료기술연구개발사업(HI18C1216) 그리고 한국연구재단(NRF-2021R1A5A8029876)(NRF-2020R1I1A1A01074256) 지원에 의하여 이루어진 것임.

References

  1. G. Briganti, and O. Le Moine, "Artificial Intelligence in Medicine: Today and Tomorrow," Frontiers in Medicine, Vol. 7, Article. 27, Feb. 2020. doi: 10.3389/fmed.2020.00027. PMID: 32118012; PMCID: PMC7012990.
  2. B. S. Tae, U. Balpukov, S. Y. Cho, and C. W. Jeong, "Elevenyear Cumulative Incidence and Estimated Lifetime Prevalence of Urolithiasis in Korea: A National Health Insurance Service- National Sample Cohort Based Study," Journal of Korean Medical Science, Vol.33, No.2, pp.e13. Jan. 2018. https://doi.org/10.3346/jkms.2018.33.e13
  3. H.-J. Kim and T.-J. Ji, "Usability Evaluation of Applied Lowdose CT When Examining Urinary Calculus Using Computed Tomography," 2017 The Journal of the Korea Association, Vol.17, Iss.6, pp.81-85, 2017.
  4. Flask [Internet], https://flask.palletsprojects.com/en/1.1.x/
  5. R. S. Olson et al., "A system for accessible artificial intelligence," in Genetic Programming Theory and Practice XV. Springer, pp.121-134, 2018.
  6. C. Olston et al., "TensorFlow-Serving: Flexible, high-performance ML serving," in 31st Conference on Neural Information Processing Systems, 2017.
  7. D. Crankshaw et al., "Clipper: A low-latency online prediction serving system," in 14th USENIX Symposium on Networked Systems Design and Implementation (NSDI), 2017, pp.613-627.
  8. Amazon SageMaker [Internet], https://docs.aws.amazon.com/sagemaker/latest/dg/whatis.html. Accessed October 14, 2018.
  9. R. Chard, et al., "DLHub: Model and Data Serving for Science," In 2019 IEEE International Parallel and Distributed Processing Symposium (IPDPS), IEEE, pp.283-292. 2019.
  10. NGINX Whitepaper [Internet], https://www.nginx.com/blog/nx_info_types/whitepaper/
  11. Gunicorn Document [Internet], https://docs.gunicorn.org/en/stable/
  12. Y. Huo, "Stochastic tissue window normalization of deep learning on computed tomography," Journal of Medical Imaging, Vol.6, No.4, pp.044005, 2019.
  13. T. D. Webb, et al., "Measurements of the Relationship Between CT Hounsfield Units and Acoustic Velocity and How It Changes With Photon Energy and Reconstruction Method," IEEE Transactions on Ultrasonics, Ferroelectrics, Vol.65, Iss.7, pp.1111-1124, Apr. 2018. https://doi.org/10.1109/TUFFC.2018.2827899