DOI QR코드

DOI QR Code

A Study on the Application of Deep Learning Model by Using ACR Phantom in CT Quality Control

CT 정도관리에서 ACR 팬텀을 이용한 딥러닝 모델 적용에 관한 연구

  • Eun-Been Choi (Department of Radiological Science, Eulji University) ;
  • Si-On Kim (Department of Radiological Science, Eulji University) ;
  • Seung-Won Choi (Department of Radiological Science, Eulji University) ;
  • Jae-Hee Kim (Department of Radiological Science, Eulji University) ;
  • Young-Kyun Kim (Department of Radiology, Samsung Medical Center) ;
  • Dong-Kyun Han (Department of Radiological Science, Eulji University)
  • 최은빈 (을지대학교 방사선학과) ;
  • 김시온 (을지대학교 방사선학과) ;
  • 최승원 (을지대학교 방사선학과) ;
  • 김재희 (을지대학교 방사선학과) ;
  • 김영균 (삼성서울병원 영상의학과 ) ;
  • 한동균 (을지대학교 방사선학과)
  • Received : 2023.11.17
  • Accepted : 2023.12.05
  • Published : 2023.12.31

Abstract

This study aimed to implement a deep learning model that can perform quantitative quality control through ACTS software used for quantitative evaluation of ACR phantom in CT quality control and evaluate its usefulness. By changing the scanning conditions, images of three modules of the ACR phantom's slice thickness (ST), low contrast resolution (LC), and high contrast resolution (HC) were obtained and classified as ACTS software. The deep learning model used ResNet18, implementing three models in which ST, HC, and LC were learned with epoch 50 and an integrated model in which three modules were learned with Epoch 10, 30, and 50 at once. The performance of each model was evaluated through Accuracy and Loss. When comparing and evaluating the accuracy and loss function values of the deep learning models by ST, LC, and HC modules, the Accuracy and Loss of the HC model were the best with 100% and 0.0081, and in the integrated model according to the Epoch value, Accuracy and Loss with epoch 50 were the best with 96.29% and 0.1856. This paper showed that quantitative quality control is possible through a deep learning model, and it can be used as a basis and evidence for applying deep learning to the CT quality control.

Keywords

Acknowledgement

This research was supported by 2023 Eulji University Innovation Support Project grant funded.

References

  1. Lee KB. Application of quantitative assessment method based on digital image processing technique for CT quality control [master's thesis]. Eulji University; 2018. Retrieved from http://www.riss.kr/link?id=T15020072 
  2. Hwang HS, Kim DH, Kim HC. Study on the application of artificial intelligence model for CT quality control. Journal of Biomedical Engineering Research. 2023;44(3):182-9. DOI: https://doi.org/10.9718/JBER.2023.44.3.182 
  3. Kiss N, Cz'uni L. Mushroom image classification with CNNs: A case-study of different learning strategies. International Symposium on Image and Signal Processing and Analysis(ISPA); 2021. DOI: https://doi.org/10.1109/ISPA52656.2021.9552053 
  4. Litjens G, Kooi T, Bejnordi BE, Setio AAA, Ciompi F, Ghafoorian M, et al. A survey on deep learning in medical image analysis. Medical Image Analysis. 2017;42:60-88. DOI: https://doi.org/10.1016/j.media.2017.07.005 
  5. Patro VM, Patra RM. Augmenting weighted average with confusion matrix to enhance classification accuracy. Transactions on Machine Learning and Artificial Intelligence. 2014;2(4):77-91. DOI: https://doi.org/10.14738/tmlai.24.328 
  6. Kang MJ. Comparison of gradient descent for deep learning. Journal of the Korea Academia-Industrial Cooperation Society. 2020;21(2):189-94. DOI: https://doi.org/10.5762/KAIS.2020.21.2.189 
  7. Cho PK. Computed tomography and quality management. Journal of the Korean Society of Radiology. 2020; 14(3):221-33. DOI: https://doi.org/10.7742/jksr.2020.14.3.221 
  8. Chang KH. Evaluation of the image quality on computed tomography with AAPM, ACR, catphan phantom [master's thesis]. Far East University; 2014. Retrieved from https://scholarworks.bwise.kr/skku/handle/2021.sw.skku/10152 
  9. Maritza A, Emilie T, Stephen D, Parker W. Using the ACR CT accreditation phantom for routine image quality assurance on both CT and CBCT imaging systems in a radiotherapy environment. Journal of Applied Clinical Medical Physics. 2014;15(4):226-39. DOI: https://doi.org/10.1120/jacmp.v15i4.4835 
  10. Mansour Z, Mokhtar A, Sarhan A, Ahmed MT, El-Diasty T. Quality control of CT image using American College of Radiology (ACR) phantom. The Egyptian Journal of Radiology and Nuclear Medicine. 2016;47(4):1665-71. DOI: https://doi.org/10.1120/jacmp.v15i4.4835 
  11. Yoo MS. The study of rationalization between dose and image quality using ACR phantom with abdominal computed tomography(CT) [master's thesis]. Eulji University; 2018. Retrieved from https://eulji.dcollection.net/srch/srchDetail/200000008500 
  12. Kim JH, Kim MC, Nam YC. Estimation of low contrast resolution in multi-detector CT: Comparison of CATPHAN, ACR phantom and AAPM phantom. Journal of Korean Society of Computed Tomographic Technology. 2009;120-6. Retrieved from http://www.riss.kr/link?id=A105245246  105245246
  13. Lee KS. An evaluation for spatial resolution, using a single target on a medical image. Journal of Radiological Science and Technology. 2016;39(4):631-6. DOI: https://doi.org/10.17946/JRST.2016.39.4.20 
  14. Jang KJ, Kweon DC. Case study of quality assurance for MDCT image quality evaluation using AAPM CT performance phantom. The Journal of the Korea Contents Association. 2007;7(7):114-23. DOI: https://doi.org/10.5392/JKCA.2007.7.7.114 
  15. Shin SK, Lee SH, Han HH. A study on residual U-Net for semantic segmentation based on deep learning. Journal of Digital Convergence. 2021;19(6):251-7. DOI: https://doi.org/10.14400/JDC.2021.19.6.251 
  16. Ying X. An overview of overfitting and its solutions. Journal of Physics: Conference Series. 2019;1168(2). DOI: https://doi.org/10.1088/1742-6596/1168/2/022022