[ Article ]
Journal of the KNST - Vol. 6, No. 2, pp.140-145
ISSN: 2635-4926
(Print)
Print publication date 30 Jun 2023
Received 31 May 2023
Revised 13 Jun 2023
Accepted 29 Jun 2023
상륙작전 시 대규모 외상환자의 신속 분류를 위한 CNN 모델
Use of CNN for Rapid Classification of Massive Trauma Patients During Amphibious Operations
Correspondence to: *Dooyoung Kim Dept. of Artificial Intelligence, Republic of Korea Naval Academy 1 Jungwon-ro, Jinhae-gu, Changwon-si, Gyungsangnam-do, 51704, Republic of Korea Tel: +82-2-907-5246 E-mail: dykim07@navy.ac.kr
© 2023 Korea Society for Naval Science & Technology
초록
본 논문에서는 대규모 외상환자가 발생하는 상륙작전에서 신속한 중증도 분류를 지원하기 위한 CNN 기반의 딥러닝 모델을 제안한다. 먼저 외상환자 학습데이터를 생성하기 위해 Unity 기반의 가상환경을 구축하여 다양한 환자 이미지를 생성하였다. CNN을 이용한 이미지 기반 환자 분류 모델을 설계하였고, 앞서 생성한 가상 이미지를 이용하여 학습을 진행하였다. 실험결과 97 %의 정확도로 부상 부위를 분류하였다.
Abstract
In this paper, we propose a CNN-based deep learning model for rapid classification in amphibious operations where massive trauma patients occur. First, we built a virtual environment based on Unity to generate various patient images for trauma patient training data. We designed an image-based classification model using CNN and proceeded with training using the generated images. As a result of the experiment, we classified the injury site with an accuracy of 97 %.
Keywords:
Triage, Amphibious Operation, Convolutional Neural Network, Multi-label Classification키워드:
중증도 분류, 상륙작전, 합성곱 신경망, 복수 라벨 분류References
- 합동·연합작전 군사용어사전, 합동참모본부, 2004, p. 224.
- 윤지한, “재난현장에서의 다수사상자 중증도 분류에 대한 인식과 전문교육훈련의 필요성에 관한 연구,” 국내석사학위논문 경북대학교 수사과학대학원, 2014, 대구.
- 소방청, 재난현장표준작전절차(SOP), 소방청, pp. 151-152.
- Robert A. De Lorenzo, MC FS USA, Improving Combat Casualty Care and Field Medicine: Focus on the Military Medic, Military Medicine, Vol. 162, No. 4, April 1997, pp. 268–272. [https://doi.org/10.1093/milmed/162.4.268]
- Su-Rak Son, Byung-Kwan Lee, Son-Kweon Sim, Yi-Na Jeong, “A Design of Passenger Detection and Sharing System(PDSS) to support the Driving (Decision) of an Autonomous Vehicles,” Journal of Korea Institute of Information, Electronics, and Communication Technology, Vol. 13, No. 2, 2020, pp. 138-144.
- Robert A. De Lorenzo, MC FS USA, Improving Combat Casualty Care and Field Medicine: Focus on the Military Medic, Military Medicine, Vol. 162, No. 4, April 1997, pp. 268–272. [https://doi.org/10.1093/milmed/162.4.268]
- Unity game engine, www.unity.com
- 대한민국 법률 제19421호, 「의료법」 제 19조(정보누설금지), 대한민국 국가법령정보센터.
- 대한민국 법률 제16930호, 「개인정보보호법」 제15조(개인정보의수집·이용), 대한민국 국가법령정보센터.
- J. Nagi et al., “Max-pooling convolutional neural networks for vision-based hand gesture recognition,” 2011 IEEE International Conference on Signal and Image Processing Applications (ICSIPA), Kuala Lumpur, Malaysia, 2011, pp. 342-347. [https://doi.org/10.1109/ICSIPA.2011.6144164]
- Agarap, A. F. Deep learning using rectified linear units (relu). ArXiv Preprint ArXiv:1803.08375, , 2018.
- Goodfellow, Ian, Yoshua Bengio, and Aaron Courville. Deep learning. MIT Press, 2016. p. 7.
- Glorot, Xavier, and Yoshua Bengio, “Understanding the difficulty of training deep feedforward neural networks,” Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, JMLR Workshop and Conference Proceedings, 2010.
- Kingma, Diederik P., and Jimmy Ba, “Adam: A method for stochastic optimization,” arXiv preprint arXiv:1412.6980, , 2014.
- Goodfellow, Ian, Yoshua Bengio, and Aaron Courville. Deep learning, MIT Press, 2016, p. 73.