딥러닝 모델을 이용한 CT 데이터 위조 및 탐지에 관한 연구
A Study on the Forgery and Detection of CT Data Using Deep Learning Model
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
최근에는 영상 처리 분야가 급속히 발전하고 있으며, 그중에서도 영상 합성 분야에서는 구별할 수 없는 우수한 품질의 가짜 영상이 생산되고 있다. 이는 의료 AI 분야의 데이터 부족 문제를 해결하는 방안으로 작용할 수 있다. 그러나 동시에 악의적인 의도 합성과 범죄를 유발할 수 있으며, 사회에 부정적인 영향을 미칠 수 있다. 따라서 본 논문에서는 GAN(Generative Adversarial Networks)을 사용하여 CT 데이터 생성기를 개발하고, 의료 AI 분야에서 사용 가능한 의미 있는 데이터를 생성하기 위해 합성 데이터의 품질을 평가한다. 그 후, 생성된 데이터와 실제 데이터를 사용하여 위조 감지 모델을 구성하고 성능을 평가한다. 이것은 악의적인 합성과 범죄에 사용될 수 있는 탐지기를 구축하기 위한 것이다. 마지막으로, 우리는 CT 데이터 생성모델을 통해 실제 데이터와 생성 데이터 그룹을 구성한다. 그리고 이를 활용한 위변조 탐지모델 개발을 통해 의료 AI 분야의 데이터 증강 기술의 가능성 연구와 악의적인 위변조 탐지 기술의 준비에 긍정적인 기여를 할 것으로 기대한다.
Abstract
Currently, research on deep learning models is being actively conducted, and many studies are being conducted in various fields. In the medical field, research is also being conducted using various medical data such as medical images and electronic medical records. In particular, research is underway to solve research problems in the medical field due to lack of data by using image conversion technology. However, through highly developed AI models, forged images can cause many problems. If forged images are abused in the medical field, it is a big problem that can endanger patients' lives. In this paper, the Cycle-GAN forgery model is constructed using CT images to solve and prevent problems that may arise through forgery images generated through artificial intelligence models. After that, the quality evaluation of the forged image was conducted. In addition, forgery image detection was performed through a forgery detection model composed of a Concat block, and detection performance was quantified with a classification performance evaluation index. As a result of classification performance indicators, each indicator showed a high value of more than 99%. Through this, it can be seen that the CT image forgery detection model of this paper has excellent performance. In the future, research will be conducted using various medical data as well as CT images, and the detection of forged data will be conducted in various ways.
참고문헌 (19)
- P. Pranita, "Deep learning a boon for biophotonics", Journal of biophotonics, 13(6) (2020)
- J. Li, "Insufficient data can also rock!learning to converse using smaller data with augmentation", Proceedings of the AAAI Conference on Artificial Intelligence, 33(1), pp.6698~6705 (2019)
- A. Dodge, "Using fake video technology to perpetuate intimate partner abuse" (2018), Without My Consent
- S. Tariq, "Am I a real or fake celebrity? Measuring commercial face recognition web APIs under deepfake impersonation attack", ~ (2021)
- M. Sharma, "A review of deepfake technology: an emerging AI threat", pp.605~619 (2022), Soft Computing for Security Applications
- L. P. D. T. Cahyani, "Adaptive histogram equalization to increase the percentage of face recognition", 8(12), pp.446~451 (2019)
- H. Najmul, "The retinex based improved underwater image enhancement", Multimedia Tools and Applications, 80(2), pp.1839~1857 (2021)
- O. Ronneberger, "U-net: Convolutional networks for biomedical image segmentation", International Conference on Medical image computing and computer-assisted intervention, pp.234~241 (2015)
- P. Esser, "A variational u-net for conditional appearance and shape generation", Proceedings of the IEEE conference on computer vision and pattern recognition, pp.8857~8866 (2018)
- J. Long, "Fully convolutional networks for semantic segmentation", Proceedings of the IEEE conference on computer vision and pattern recognition, pp.3431~3440 (2015)
- A. Mohammed, "Soft computing techniques: systematic multiscale models to predict the compressive strength of HVFA concrete based on mix proportions and curing times", Journal of Building Engineering, 33(101851) (2021)
- U. Sara, "Image quality assessment through FSIM, SSIM, MSE and PSNR—a comparative study", Journal of Computer and Communications, 7(3), pp.8~18 (2019)
- 송호중, "U-Net 모델을 이용한 안저 이미지 재생성에 관한 연구", 한국지식정보기술학회 논문지, 17(3), pp.435~443 (2022), 한국지식정보기술학회[DOI]
- Y. Ma, "A precision evaluation index system for remote sensing data sampling based on hexagonal discrete grids", ISPRS International Journal of Geo-Information, 10(3), pp.194~ (2021)
- D. Baab, "Longitudinal evaluation of a self‐inspection plaque index in periodontal recall patients", Journal of clinical periodontology, 13(4), pp.313~318 (1986)
- D. Chicco, "The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation", BMC genomics, 21(1), pp.1~13 (2020)
- K. Hajian-Tilaki, "Receiver operating characteristic (ROC) curve analysis for medical diagnostic test evaluation", Caspian journal of internal medicine, 4(2), pp.627~ (2013)
- J Huang, "Using AUC and accuracy in evaluating learning algorithms", IEEE Transactions on knowledge and Data Engineering, 17(3), pp.299~310 (2005)
- 서상민, "감시 카메라를 이용한 딥러닝 기반의 화재 경보 시스템", 한국지식정보기술학회 논문지, 16(3), pp.423~432 (2021), 한국지식정보기술학회[DOI]