Journal of radiological science and technology (대한방사선기술학회지:방사선기술과학)

Korean Society of Radiological Science (대한방사선과학회)
권호 표지
DOI QR코드

DOI QR Code

Enhancing CT Image Quality Using Conditional Generative Adversarial Networks for Applying Post-mortem Computed Tomography in Forensic Pathology: A Phantom Study

사후전산화단층촬영의 법의병리학 분야 활용을 위한 조건부 적대적 생성 신경망을 이용한 CT 영상의 해상도 개선: 팬텀 연구

  • Yebin Yoon (Dept. of Multidisciplinary Radiological Science, The Graduate School of Dongseo University) ;
  • Jinhaeng Heo (Forensic Medicine Div., Busan Institute, National Forensic Service) ;
  • Yeji Kim (Dept. of Multidisciplinary Radiological Science, The Graduate School of Dongseo University) ;
  • Hyejin Jo (Dept. of Multidisciplinary Radiological Science, The Graduate School of Dongseo University) ;
  • Yongsu Yoon (Dept. of Multidisciplinary Radiological Science, The Graduate School of Dongseo University)
  • 윤예빈 (동서대학교 일반대학원 융합방사선학과) ;
  • 허진행 (국립과학수사연구원 부산과학수사연구소 법의학과) ;
  • 김예지 (동서대학교 일반대학원 융합방사선학과) ;
  • 조혜진 (동서대학교 일반대학원 융합방사선학과) ;
  • 윤용수 (동서대학교 일반대학원 융합방사선학과)
  • Received : 2023.07.25
  • Accepted : 2023.08.16
  • Published : 2023.08.31
Download PDF
⟨ Previous Next ⟩

Abstract

Post-mortem computed tomography (PMCT) is commonly employed in the field of forensic pathology. PMCT was mainly performed using a whole-body scan with a wide field of view (FOV), which lead to a decrease in spatial resolution due to the increased pixel size. This study aims to evaluate the potential for developing a super-resolution model based on conditional generative adversarial networks (CGAN) to enhance the image quality of CT. 1761 low-resolution images were obtained using a whole-body scan with a wide FOV of the head phantom, and 341 high-resolution images were obtained using the appropriate FOV for the head phantom. Of the 150 paired images in the total dataset, which were divided into training set (96 paired images) and validation set (54 paired images). Data augmentation was perform to improve the effectiveness of training by implementing rotations and flips. To evaluate the performance of the proposed model, we used the Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM) and Deep Image Structure and Texture Similarity (DISTS). Obtained the PSNR, SSIM, and DISTS values of the entire image and the Medial orbital wall, the zygomatic arch, and the temporal bone, where fractures often occur during head trauma. The proposed method demonstrated improvements in values of PSNR by 13.14%, SSIM by 13.10% and DISTS by 45.45% when compared to low-resolution images. The image quality of the three areas where fractures commonly occur during head trauma has also improved compared to low-resolution images.

Keywords

Acknowledgement

This research was conducted with the support of the 2021 Academic Research Grant from Dongseo University. (DSU-20210010)

References

  1. Chi J, Sun Z, Wang H, Lyu P, Yu X, Wu C. CT image super-resolution reconstruction based on global hybrid attention. Computers in Biology and Medicine. 2022;150:106112.
  2. Lee MK, Lim CH. Survey of CT practice and collective effective dose estimation. Journal of Radiological Science and Technology. 2010;33(3):231-7.
  3. Krantz P, HoltHs S. Postmortem computed tomography in a diving fatality. Journal of Computer Assisted Tomography. 1983;7(1):132-4. https://doi.org/10.1097/00004728-198302000-00024
  4. Morishita J, Ueda Y. New solutions for automated image recognition and identification: Challenges to radiologic technology and forensic pathology. Radiological Physics and Technology. 2021;14:123-33. https://doi.org/10.1007/s12194-021-00611-9
  5. Heo JH, Yoon Y, Seo IS, Im SB, San Ko Y, Lee S, Jang SJ. The significance of evaluating sphenoid sinus fluid by postmortem computed tomography in cases of drowning. Journal of Forensic and Legal Medicine. 2023;102551.
  6. Kawazoe Y, Morishita J, Matsunobu Y, Okumura M, Shin S, Usumoto Y, Ikeda N. A simple method for semi-automatic readjustment for positioning in post-mortem head computed tomography imaging. Journal of Forensic Radiology and Imaging. 2019;16:57-64. https://doi.org/10.1016/j.jofri.2019.01.004
  7. Ross SG, Bolliger SA, Ampanozi G, Oesterhelweg L, Thali MJ, Flach PM. Postmortem CT angiography: Capabilities and limitations in traumatic and natural causes of death. Radiographics. 2014;34(3):830-46. https://doi.org/10.1148/rg.343115169
  8. Grabherr S, Egger C, Vilarino R, Campana L, Jotterand M, Dedouit F. Modern post-mortem imaging: An update on recent developments. Forensic Sciences Research. 2017;2(2):52-64. https://doi.org/10.1080/20961790.2017.1330738
  9. Gascho D, Thali MJ, Niemann T. Post-mortem computed tomography: Technical principles and recommended parameter settings for high-resolution imaging. Medicine, Science and the Law. 2018;58(1):70-82. https://doi.org/10.1177/0025802417747167
  10. Flach PM, Gascho D, Schweitzer W, Ruder TD, Berger N, Ross SG, et al. Imaging in forensic radiology: An illustrated guide for postmortem computed tomography technique and protocols. Forensic Science, Medicine, and Pathology. 2014;10:583-606. https://doi.org/10.1007/s12024-014-9555-6
  11. Kempter M, Ross S, Spendlove D, Flach PM, Preiss U, Thali MJ, Bolliger SA. Post-mortem imaging of laryngohyoid fractures in strangulation incidents: First results. Legal Medicine. 2009;11(6):267-71. https://doi.org/10.1016/j.legalmed.2009.07.005
  12. Grassberger M, Gehl A, PHschel K, Turk EE. 3D reconstruction of emergency cranial computed tomography scans as a tool in clinical forensic radiology after survived blunt head trauma-report of two cases. Forensic Science International. 2011;207(1-3):e19-23. https://doi.org/10.1016/j.forsciint.2010.11.014
  13. Sarment DP, Christensen AM. The use of cone beam computed tomography in forensic radiology. Journal of Forensic Radiology and Imaging. 2014;2(4):173-81. https://doi.org/10.1016/j.jofri.2014.09.002
  14. Thali MJ, Yen K, Schweitzer W, Vock P, Boesch C, Ozdoba C, et al. Virtopsy, a new imaging horizon in forensic pathology: Virtual autopsy by postmortem multislice computed tomography (MSCT) and magnetic resonance imaging (MRI)-A feasibility study. Journal of Forensic Sciences. 2003; 48(2):386-403. https://doi.org/10.1520/JFS2002166
  15. Hong, JY, Park SH, Jung YJ. Artificial intelligence based medical imaging: An overview. Journal of Radiological Science and Technology. 2020;43(3):195-208. https://doi.org/10.17946/JRST.2020.43.3.195
  16. Mirza M, Osindero S. Conditional generative adversarial nets. arXiv preprint arXiv; 2014:1411-784.
  17. Zijlstra F, Wilemsen K, Florkow MC, Sakkers RJ, Weinans HH, Van der Wal BCH, et al. CT synthesis from MR images for orthopedic applications in the lower arm using a conditional generative adversarial network. Medical Imaging 2019: Image Processing. 2019;10949:387-93. https://doi.org/10.1117/12.2512857
  18. Wang J, Noble JH, Dawant BM. Metal artifact reduction for the segmentation of the intra cochlear anatomy in CT images of the ear with 3D-conditional GANs. Medical Imaging Analysis. 2019;58:101553.
  19. Shin JH, Lee MJ, Park SS, Jeong WJ, You YH. Severe traumatic intraocular injuries related to blowout fractures. Journal of the Korean Society of Traumatology. 2009;22(1):97-102.
  20. Park SH, Rah SH, KIM YH. Clinical evaluation of the associated ocular injuries of orbital wall fracture patients. Journal of the Korean Ophthalmological Society. 2002:1474-81.
  21. Kim J, Hwang W. Delayed reduction of zygomatic arch fracture: Effectiveness of the Rowe Zygoma Elevator. Journal of Craniofacial Surgery. 2018;29(7):e639-40. https://doi.org/10.1097/SCS.0000000000004622
  22. Lee JH, Paik JY. Diagnosis and treatment of temporal bone trauma. Korean Journal of Otorhinolaryngology-Head and Neck Surgery. 2014;57(7):433-41. https://doi.org/10.3342/kjorl-hns.2014.57.7.433
  23. Hore A, Ziou D. Image quality metrics: PSNR vs. SSIM. In 2010 20th International Conference on Pattern Recognition, IEEE; 2010:2366-9.
  24. Ding K, Ma K, Wang S, Simoncelli EP. Image quality assessment: Unifying structure and texture similarity. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2020;44(5):2567-81. https://doi.org/10.1109/TPAMI.2020.3045810
  25. Aoki T, Ito K, Aoyama S, Kosuge E. Disaster victim identification using dental records-Experience of the Great East Japan Earthquake. In 2013 IEEE Region 10 Humanitarian Technology Conference, IEEE; 2013:57-62.
  26. Pfaeffli M, Vock P, Dirnhofer R, Braun M, Bolliger SA, Thali MJ. Post-mortem radiological CT identification based on classical ante-mortem X-ray examinations. Forensic Science International. 2007; 171(2-3):111-7. https://doi.org/10.1016/j.forsciint.2006.10.009
  27. Quatrehomme G, Fronty P, Sapanet M, GrHvin G, Bailet P, Ollier A. Identification by frontal sinus pattern in forensic anthropology. Forensic Science International. 1996;83(2):147-53. https://doi.org/10.1016/S0379-0738(96)02033-6
  28. Tsubaki S, Morishita J, Usumoto Y, Sakaguchi K, Matsunobu Y, Kawazoe Y, et al. Sex determination based on a thoracic vertebra and ribs evaluation using clinical chest radiography. Legal Medicine. 2017;27:19-24. https://doi.org/10.1016/j.legalmed.2017.06.003
  29. Kano T, Oritani S, Michiue T, Ishikawa T, Hishmat AM, Sogawa N, et al. Postmortem CT morphometry with a proposal of novel parameters for sex discrimination of the mandible using Japanese adult data. Legal Medicine. 2015;17(3):167-71. https://doi.org/10.1016/j.legalmed.2014.12.009
  30. Akhlaghi M, Bakhtavar K, Moarefdoost J, Kamali A, Rafeifar S. Frontal sinus parameters in computed tomography and sex determination. Legal Medicine. 2016;19:22-7.  https://doi.org/10.1016/j.legalmed.2016.01.008