Journal article
Forensic 3D Visualization of CT Data Using Cinematic Volume Rendering: A Preliminary Study.
- Ebert LC 1 Institute of Forensic Medicine, University of Zurich, Winterthurerstrasse 190/52, 8057 Zurich, Switzerland.
- Schweitzer W 1 Institute of Forensic Medicine, University of Zurich, Winterthurerstrasse 190/52, 8057 Zurich, Switzerland.
- Gascho D 1 Institute of Forensic Medicine, University of Zurich, Winterthurerstrasse 190/52, 8057 Zurich, Switzerland.
- Ruder TD 1 Institute of Forensic Medicine, University of Zurich, Winterthurerstrasse 190/52, 8057 Zurich, Switzerland.
- Flach PM 1 Institute of Forensic Medicine, University of Zurich, Winterthurerstrasse 190/52, 8057 Zurich, Switzerland.
- Thali MJ 1 Institute of Forensic Medicine, University of Zurich, Winterthurerstrasse 190/52, 8057 Zurich, Switzerland.
- Ampanozi G 1 Institute of Forensic Medicine, University of Zurich, Winterthurerstrasse 190/52, 8057 Zurich, Switzerland.
- 2016-11-09
Published in:
- AJR. American journal of roentgenology. - 2017
3D visualization
cinematic rendering
forensic imaging
postmortem CT
volume rendering
Autopsy
Forensic Medicine
Imaging, Three-Dimensional
Pilot Projects
Tomography, X-Ray Computed
User-Computer Interface
Video Recording
English
OBJECTIVE
The 3D volume-rendering technique (VRT) is commonly used in forensic radiology. Its main function is to explain medical findings to state attorneys, judges, or police representatives. New visualization algorithms permit the generation of almost photorealistic volume renderings of CT datasets. The objective of this study is to present and compare a variety of radiologic findings to illustrate the differences between and the advantages and limitations of the current VRT and the physically based cinematic rendering technique (CRT).
MATERIALS AND METHODS
Seventy volunteers were shown VRT and CRT reconstructions of 10 different cases. They were asked to mark the findings on the images and rate them in terms of realism and understandability.
RESULTS
A total of 48 of the 70 questionnaires were returned and included in the analysis. On the basis of most of the findings presented, CRT appears to be equal or superior to VRT with respect to the realism and understandability of the visualized findings. Overall, in terms of realism, the difference between the techniques was statistically significant (p < 0.05). Most participants perceived the CRT findings to be more understandable than the VRT findings, but that difference was not statistically significant (p > 0.05).
CONCLUSION
CRT, which is similar to conventional VRT, is not primarily intended for diagnostic radiologic image analysis, and therefore it should be used primarily as a tool to deliver visual information in the form of radiologic image reports. Using CRT for forensic visualization might have advantages over using VRT if conveying a high degree of visual realism is of importance. Most of the shortcomings of CRT have to do with the software being an early prototype.
The 3D volume-rendering technique (VRT) is commonly used in forensic radiology. Its main function is to explain medical findings to state attorneys, judges, or police representatives. New visualization algorithms permit the generation of almost photorealistic volume renderings of CT datasets. The objective of this study is to present and compare a variety of radiologic findings to illustrate the differences between and the advantages and limitations of the current VRT and the physically based cinematic rendering technique (CRT).
MATERIALS AND METHODS
Seventy volunteers were shown VRT and CRT reconstructions of 10 different cases. They were asked to mark the findings on the images and rate them in terms of realism and understandability.
RESULTS
A total of 48 of the 70 questionnaires were returned and included in the analysis. On the basis of most of the findings presented, CRT appears to be equal or superior to VRT with respect to the realism and understandability of the visualized findings. Overall, in terms of realism, the difference between the techniques was statistically significant (p < 0.05). Most participants perceived the CRT findings to be more understandable than the VRT findings, but that difference was not statistically significant (p > 0.05).
CONCLUSION
CRT, which is similar to conventional VRT, is not primarily intended for diagnostic radiologic image analysis, and therefore it should be used primarily as a tool to deliver visual information in the form of radiologic image reports. Using CRT for forensic visualization might have advantages over using VRT if conveying a high degree of visual realism is of importance. Most of the shortcomings of CRT have to do with the software being an early prototype.
- Language
-
- English
- Open access status
- closed
- Identifiers
-
- DOI 10.2214/AJR.16.16499
- PMID 27824494
- Persistent URL
- https://sonar.ch/global/documents/259421
Statistics
Document views: 20
File downloads: