대한영상의학회지 (Journal of the Korean Society of Radiology)

  • 제83권6호
  • /
  • Pages.1208-1218
  • /
  • 2022
  • /
  • 1738-2637(pISSN)
  • /
  • 2288-2928(eISSN)
대한영상의학회 (The Korean Society of Radiology)
권호 표지
DOI QR코드

DOI QR Code

의료 인공지능에서의 멀티 태스크 러닝의 이해와 활용

Understanding and Application of Multi-Task Learning in Medical Artificial Intelligence

  • 김영재 (가천대학교 의용생체공학과) ;
  • 김광기 (가천대학교 의용생체공학과)
  • Young Jae Kim (Department of Biomedical Engineering, Gachon University) ;
  • Kwang Gi Kim (Department of Biomedical Engineering, Gachon University)
  • 투고 : 2022.11.14
  • 심사 : 2022.11.24
  • 발행 : 2022.11.01
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

최근, 의료 분야에서 인공지능은 많은 발전을 통해 다양한 분야로 확장하며 활용되고 있다. 하지만 대부분의 인공지능 기술들은 하나의 모델이 하나의 태스크만을 수행할 수 있도록 개발되고 있으며, 이는 의사들의 복잡한 판독 과정을 인공지능으로 설계하는데 한계로 작용한다. 멀티 태스크 러닝은 이러한 한계를 극복하기 위한 최적의 방안으로 알려져 있다. 다양한 태스크들을 동시에 하나의 모델로 학습함으로써, 효율적이고 일반화에 유리한 모델을 만들수 있다. 본 종설에서는 멀티 태스크 러닝에 대한 개념과 종류, 유사 개념 등에 대해 알아보고, 연구 사례들을 통해 의료 분야에서의 멀티 태스크 러닝의 활용 현황과 향후 가능성을 살펴보고자 한다.

In the medical field, artificial intelligence has been used in various ways with many developments. However, most artificial intelligence technologies are developed so that one model can perform only one task, which is a limitation in designing the complex reading process of doctors with artificial intelligence. Multi-task learning is an optimal way to overcome the limitations of single-task learning methods. Multi-task learning can create a model that is efficient and advantageous for generalization by simultaneously integrating various tasks into one model. This study investigated the concepts, types, and similar concepts as multi-task learning, and examined the status and future possibilities of multi-task learning in the medical research.

키워드

과제정보

This work was supported by the Technology Innovation Program (K_G012001187801, "Development of Diagnostic Medical Devices with Artificial intelligence Based Image Analysis Technology") funded By the Ministry of Trade, Industry & Energy (MOTIE, Korea), and by the Gachon Gil Medical Center (FRD2020-18).

참고문헌

  1. Sharma N, Sharma R, Jindal N. Machine learning and deep learning applications-a vision. Global Transitions Proceedings 2021;2:24-28 
  2. Montagnon E, Cerny M, Cadrin-Chenevert A, Hamilton V, Derennes T, Ilinca A, et al. Deep learning workflow in radiology: a primer. Insights Imaging 2020;11:22 
  3. Li Y, Sixou B, Peyrin F. A review of the deep learning methods for medical images super resolution problems. IRBM 2021;42:120-133 
  4. Wang T, Lei Y, Fu Y, Wynne JF, Curran WJ, Liu T, et al. A review on medical imaging synthesis using deep learning and its clinical applications. J Appl Clin Med Phys 2021;22:11-36 
  5. Haskins G, Kruger U, Yan P. Deep learning in medical image registration: a survey. Mach Vis Appl 2020;31:8 
  6. Zhang HM, Dong B. A review on deep learning in medical image reconstruction. J Oper Res Soc China 2020;8:311-340 
  7. Chan HP, Samala RK, Hadjiiski LM, Zhou C. Deep learning in medical image analysis. In Lee G, Fujita H. eds. Deep learning in medical image analysis: challenges and applications. Cham: Springer 2020:3-21 
  8. Chan HP, Hadjiiski LM, Samala RK. Computer-aided diagnosis in the era of deep learning. Med Phys 2020; 47:e218-e227 
  9. Zhang Y, Yang Q. A survey on multi-task learning. IEEE Transactions on Knowledge and Data Engineering 2022;34:5586-5609 
  10. Zhang Y, Yang Q. An overview of multi-task learning. Natl Sci Rev 2017;5:30-43 
  11. Ruder S. An overview of multi-task learning in deep neural networks. arXiv [Preprint]. 2017 [cited November 2, 2022]. Available at: https://doi.org/10.48550/arXiv.1706.05098 
  12. Chen Y, Zhao D, Lv L, Zhang Q. Multi-task learning for dangerous object detection in autonomous driving. Inf Sci 2018;432:559-571 
  13. Vandenhende S, Georgoulis S, Proesmans M, Dai D, Van Gool L. Revisiting multi-task learning in the deep learning era. arXiv [Preprint]. 2004 [cited November 2, 2022]. Available at: http://www.esat.kuleuven.be/~konijn/publications/2020/vandenhende.pdf 
  14. Liu T, Tao D, Song M, Maybank SJ. Algorithm-dependent generalization bounds for multi-task learning. IEEE Trans Pattern Anal Mach Intell 2017;39:227-241 
  15. Vafaeikia P, Namdar K, Khalvati F. A brief review of deep multi-task learning and auxiliary task learning. arXiv [Preprint]. 2020 [cited November 2, 2022]. Available at: https://doi.org/10.48550/arXiv.2007.01126. 
  16. Hinton G, Vinyals O, Dean J. Distilling the knowledge in a neural network. arXiv [Preprint]. 2015 [cited November 2, 2022]. Available at: https://doi.org/10.48550/arXiv.1503.02531 
  17. Long M, Cao Z, Wang J, Yu PS. Learning multiple tasks with multilinear relationship networks. arXiv [Preprint]. 2017 [cited November 2, 2022]. Available at: https://doi.org/10.48550/arXiv.1506.02117 
  18. Lu Y, Kumar A, Zhai S, Cheng Y, Javidi T, Feris R. Fully-adaptive feature sharing in multi-task networks with applications in person attribute classification. Proceedings of the IEEE conference on Computer Vision and Pattern Recognition (CVPR); 2017 July 22-25; Honolulu, HI, USA: IEEE; 2017:5334-5343 
  19. Misra I, Shrivastava A, Gupta A, Hebert M. Cross-stitch networks for multi-task learning. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR); 2016 June 27-30; Las Vegas, NV, USA: IEEE; 2016:3994-4003 
  20. Saha B, Gupta S, Phung D, Venkatesh S. Multiple task transfer learning with small sample sizes. Knowl Inf Syst 2016;46:315-342 
  21. Pilault J, Elhattami A, Pal C. Conditionally adaptive multi-task learning: improving transfer learning in NLP using fewer parameters & less data. arXiv [Preprint]. 2020 [cited November 2, 2022]. Available at: https://doi. org/10.48550/arXiv.2009.09139 
  22. Weiss K, Khoshgoftaar TM, Wang D. A survey of transfer learning. J Big Data 2016;3:9 
  23. Kornblith S, Shlens J, Le QV. Do better imagenet models transfer better? Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR); 2019 June 16-20; Long Beach, CA, USA: IEEE; 2019:2661-2671 
  24. Vrbancic G, Podgorelec V. Transfer learning with adaptive fine-tuning. IEEE Access 2020;8:196197-196211 
  25. Zhang ML, Zhou ZH. A review on multi-label learning algorithms. IEEE Trans Knowl Data Eng 2014;26:1819- 1837 
  26. Gao BB, Zhou HY. Learning to discover multi-class attentional regions for multi-label image recognition. IEEE Trans Image Process 2021;30:5920-5932 
  27. Chaichulee S, Villarroel M, Jorge J, Arteta C, Green G, McCormick K, et al. Multi-task convolutional neural network for patient detection and skin segmentation in continuous non-contact vital sign monitoring. Proceedings of the IEEE International Conference on Automatic Face & Gesture Recognition (FG 2017); 2017 May 30-June 3; Washington, DC, USA: IEEE; 2017:266-272 
  28. Li X, Hu X, Yu L, Zhu L, Fu CW, Heng PA. CANet: cross-disease attention network for joint diabetic retinopathy and diabetic macular edema grading. IEEE Trans Med Imaging 2020;39:1483-1493 
  29. He T, Guo J, Wang J, Xu X, Yi Z. Multi-task learning for the segmentation of thoracic organs at risk in CT images. Proceedings of the IEEE International Symposium of Biomedical imaging (ISBI); 2019 April 8-11; Venice, Italy: IEEE; 2019:10-13 
  30. Zhai P, Tao Y, Chen H, Cai T, Li J. Multi-task learning for lung nodule classification on chest CT. IEEE Access 2020;8:180317-180327 
  31. Gao F, Yoon H, Wu T, Chu X. A feature transfer enabled multi-task deep learning model on medical imaging. Expert Syst Appl 2020;143:112957 
  32. Amyar A, Modzelewski R, Li H, Ruan S. Multi-task deep learning based CT imaging analysis for COVID-19 pneumonia: classification and segmentation. Comput Biol Med 2020;126:104037