익사 사례에서 Postmortem CT 영상을 활용한 U-Net 기반 나비굴 내 액체 자동 분할: 타당성 연구
U-Net-Based Automatic Segmentation of Sphenoid Sinus Fluid in Drowning Cases Using Postmortem CT Images: A Feasibility Study
0
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
Detecting sphenoid sinus fluid (SSF) is an additional finding in autopsies for diagnosing drowning. SSF can provide additional forensic evidence through laboratory tests such as diatom and electrolyte analyses. If drowning is suspected, accurately assessing the presence and volume of SSF during an autopsy is crucial. Utilizing postmortem computed tomography (PMCT) images could aid in accurately sampling SSF. Accurately segmenting the region of interest is essential for volume analysis using computed tomography images. However, manual segmentation techniques are labor-intensive and time-consuming, and their success depends on the experience of the observer. Therefore, this study aimed to develop a U-Net–based deep learning model for the automatic segmentation of SSF in drowning cases using PMCT images and to evaluate the performance of the model. We retrospectively reviewed 34 drowning cases in which both PMCT scans and forensic autopsies were performed at our institution. The U-Net architecture of deep learning was used for automatic segmentation. The proposed model achieved the Dice similarity coefficient (DSC) and Intersection over Union (IoU) of a maximum of 95.85% and 92.03%, a minimum of 0% and 0%, and an average of 77.15% and 67.18%, respectively. Although the average DSC and IoU did not show high similarity, this study showed that PMCT images can be used for automatic segmentation of SSF in drowning cases, which could improve the performance with sufficient dataset acquisition and further model training.
Abstract
Detecting sphenoid sinus fluid (SSF) is an additional finding in autopsies for diagnosing drowning. SSF can provide additional forensic evidence through laboratory tests such as diatom and electrolyte analyses. If drowning is suspected, accurately assessing the presence and volume of SSF during an autopsy is crucial. Utilizing postmortem computed tomography (PMCT) images could aid in accurately sampling SSF. Accurately segmenting the region of interest is essential for volume analysis using computed tomography images. However, manual segmentation techniques are labor-intensive and time-consuming, and their success depends on the experience of the observer. Therefore, this study aimed to develop a U-Net–based deep learning model for the automatic segmentation of SSF in drowning cases using PMCT images and to evaluate the performance of the model. We retrospectively reviewed 34 drowning cases in which both PMCT scans and forensic autopsies were performed at our institution. The U-Net architecture of deep learning was used for automatic segmentation. The proposed model achieved the Dice similarity coefficient (DSC) and Intersection over Union (IoU) of a maximum of 95.85% and 92.03%, a minimum of 0% and 0%, and an average of 77.15% and 67.18%, respectively. Although the average DSC and IoU did not show high similarity, this study showed that PMCT images can be used for automatic segmentation of SSF in drowning cases, which could improve the performance with sufficient dataset acquisition and further model training.
참고문헌 (12)
- Piette MH, "Drowning : still a difficult autopsy diagnosis", Forensic Sci Int, 163, pp.1~9 (2006)
- Lin CY, "Diatomological investigation in sphenoid sinus fluid and lung tissue from cases of suspected drowning", Forensic Sci Int, 244, pp.111~115 (2014)
- Na JY, "Evaluation of sodium and chloride biochemical tests in drowning autopsy cases", Korean J Leg Med, 43, pp.129~137 (2019)
- Lee SH, "The significance of fluid in the sphenoid sinuses in death by drowning", Korean J Leg Med, 37, pp.129~133 (2013)
- Heo JH, "The significance of evaluating sphenoid sinus fluid by postmortem computed tomography in cases of drowning", J Forensic Leg Med, 97, pp.102551~ (2023)
- Zhao W, "MSS U-Net : 3D segmentation of kidneys and tumors from CT images with a multi-scale supervised U-Net", Inform Med Unlocked, 19, pp.100357~ (2020)
- Pham CV, "Age estimation based on 3D post-mortem computed tomography images of mandible and femur using convolutional neural networks", PLoS One, 16, pp.e0251388~ (2021)
- Dobay A, "Potential use of deep learning techniques for postmortem imaging", Forensic Sci Med Pathol, 16, pp.671~679 (2020)
- Ronneberger O, "Medical Image Computing and Computer-Assisted Intervention - MICCAI 2015", pp.234~241 (2015), Springer
- Vaid S, "Normal anatomy and anatomic variants of the paranasal sinuses on computed tomography", Neuroimaging Clin N Am, 25, pp.527~548 (2015)
- Lundemose SB, "Exact volumetric determination of fluid in the paranasal sinuses after drowning", J Forensic Radiol Imaging, 3, pp.111~116 (2015)
- Lo Re G, "Role of virtopsy in the post-mortem diagnosis of drowning", Radiol Med, 120, pp.304~308 (2015)