DOI QR코드

DOI QR Code

A Study on the Use of Contrast Agent and the Improvement of Body Part Classification Performance through Deep Learning-Based CT Scan Reconstruction

딥러닝 기반 CT 스캔 재구성을 통한 조영제 사용 및 신체 부위 분류 성능 향상 연구

  • Seongwon Na (Biomedical Research Center, Asan Institute for Life Sciences, Asan Medical Center) ;
  • Yousun Ko (Department of Radiology and Research Institute of Radiology, Asan Medical Center, University of Ulsan College of Medicine) ;
  • Kyung Won Kim (Department of Radiology and Research Institute of Radiology, Asan Medical Center)
  • 나성원 (서울아산병원 의생명연구소) ;
  • 고유선 (울산대학교 의과대학 의학과) ;
  • 김경원 (서울아산병원 영상의학과)
  • Received : 2023.03.31
  • Accepted : 2023.04.25
  • Published : 2023.05.30

Abstract

Unstandardized medical data collection and management are still being conducted manually, and studies are being conducted to classify CT data using deep learning to solve this problem. However, most studies are developing models based only on the axial plane, which is a basic CT slice. Because CT images depict only human structures unlike general images, reconstructing CT scans alone can provide richer physical features. This study seeks to find ways to achieve higher performance through various methods of converting CT scan to 2D as well as axial planes. The training used 1042 CT scans from five body parts and collected 179 test sets and 448 with external datasets for model evaluation. To develop a deep learning model, we used InceptionResNetV2 pre-trained with ImageNet as a backbone and re-trained the entire layer of the model. As a result of the experiment, the reconstruction data model achieved 99.33% in body part classification, 1.12% higher than the axial model, and the axial model was higher only in brain and neck in contrast classification. In conclusion, it was possible to achieve more accurate performance when learning with data that shows better anatomical features than when trained with axial slice alone.

표준화되지 않은 의료 데이터 수집 및 관리는 여전히 수동으로 진행되고 있어, 이 문제를 해결하기 위해 딥 러닝을 사용해 CT 데이터를 분류하는 연구들이 진행되고 있다. 하지만 대부분 연구에서는 기본적인 CT slice인 axial 평면만을 기반으로 모델을 개발하고 있다. CT 영상은 일반 이미지와 다르게 인체 구조만 묘사하기 때문에 CT scan을 재구성하는 것만으로도 더 풍부한 신체적 특징을 나타낼 수 있다. 이 연구는 axial 평면뿐만 아니라 CT 데이터를 2D로 변환하는 여러가지 방법들을 통해 보다 높은 성능을 달성할 수 있는 방법을 찾고자 한다. 훈련은 5가지 부위의 CT 스캔 1042개를 사용했고, 모델 평가를 위해 테스트셋 179개, 외부 데이터셋으로 448개를 수집했다. 딥러닝 모델 개발을 위해 ImageNet으로 사전 학습된 InceptionResNetV2를 백본으로 사용하였으며, 모델의 전체 레이어를 재 학습했다. 실험결과 신체 부위 분류에서는 재구성 데이터 모델이 99.33%를 달성하며 axial 모델보다 1.12% 더 높았고, 조영제 분류에서는 brain과 neck에서만 axial모델이 높았다. 결론적으로 axial slice로만 훈련했을 때 보다 해부학적 특징이 잘 나타나는 데이터로 학습했을 때 더 정확한 성능 달성이 가능했다.

Keywords

Acknowledgement

This study was supported by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (HI18C1216).

References

  1. Russakovsky, Olga, et al. "Imagenet large scale visual recognition challenge." International journal of computer vision 115 (2015): 211-252. doi: https://doi.org/10.1007/s11263-015-0816-y
  2. Choy, Garry, et al. "Current applications and future impact of machine learning in radiology." Radiology 288.2 (2018): 318-328. doi: https://doi.org/10.1148/radiol.2018171820.
  3. Allen Jr, Bibb, et al. "A road map for translational research on artificial intelligence in medical imaging: from the 2018 National Institutes of Health/RSNA/ACR/The Academy Workshop." Journal of the American College of Radiology 16.9 (2019): 1179-1189. doi: https://doi.org/10.1016/j.jacr.2019.04.014.
  4. Gauriau, Romane, et al. "Using DICOM metadata for radiological image series categorization: a feasibility study on large clinical brain MRI datasets." Journal of digital imaging 33 (2020): 747-762. doi: https://doi.org/10.1007/s10278-019-00308-x.
  5. Cho, Junghwan, et al. "How much data is needed to train a medical image deep learning system to achieve necessary high accuracy?." arXiv preprint arXiv:1511.06348 (2015). https://doi.org/10.48550/arXiv.1511.06348
  6. Roth, Holger R., et al. "Anatomy-specific classification of medical images using deep convolutional nets." 2015 IEEE 12th international symposium on biomedical imaging (ISBI). IEEE, 2015. doi: https://doi.org/10.1109/ISBI.2015.7163826
  7. Bae, Kyongtae T. "Intravenous contrast medium administration and scan timing at CT: considerations and approaches." Radiology 256.1 (2010): 32-61. doi: https://doi.org/10.1148/radiol.10090908.
  8. Hamlin, Derek J., F. A. Burgener, and J. B. Beecham. "CT of intramural endometrial carcinoma: contrast enhancement is essential." American Journal of Roentgenology 137.3 (1981): 551-554. doi: https://doi.org/10.2214/ajr.137.3.551.
  9. Sugimori, Hiroyuki. "Classification of computed tomography images in different slice positions using deep learning." Journal of healthcare engineering 2018 (2018). doi: https://doi.org/10.1155/2018/1753480.
  10. Philbrick, Kenneth A., et al. "What does deep learning see? Insights from a classifier trained to predict contrast enhancement phase from CT images." American Journal of Roentgenology 211.6 (2018): 1184-1193. doi: https://doi.org/10.2214/ajr.18.20331.
  11. Szegedy, Christian, et al. "Inception-v4, inception-resnet and the impact of residual connections on learning." Proceedings of the AAAI conference on artificial intelligence. Vol. 31. No. 1. 2017. doi: https://doi.org/10.1609/aaai.v31i1.11231