Comparison of 3D CT VR and GIR images
The 3D effect, the smoothness of the skin surface, and for a detailed assessment of stab wounds, GIR images were superior. We found that the diagnostic value of 3D CT GIR image was similar to macroscopic photographs.
As GIR technique simulates the propagation and interaction of light rays as they pass through volumetric data3,6,7,8,9,10,11,12,13,14, it improves the 3D effect, depth perception, and shows subtle irregularities of the skin surface. VR, on the other hand, applies a local lighting model to volumetric data12. Among 3 small, shallow stab wounds that were not detectable on VR images, 2 were displayed on GIR images. VR also failed to distinguish between two adjacent stab wounds, GIR did not (see Fig. 3). Our findings indicate that GIR yields a more photorealistic representation of 3D images than the conventional VR technique.
Although the utility of GIR application to 3D CT in the medical field has been reported5,6,7,8,9, only few studies assessed the diagnostic value of GIR technique6. Ours is the first documentation that GIR is superior to conventional VR at the postmortem investigation of stabbing victims.
Advantage of GIR in the assessment of stab wounds
The detailed inspection of stab wounds is crucial for the identification of the stabbing instrument, the applied force, the stab trajectory of the weapon, and the position of the victim and perpetrator at the time of the assault17. The application of GIR to 3D CT imaging facilitates the detailed assessment of stab wounds and yields images that are comparable to macroscopic photographs.
For the assessment of stab wounds, 3D CT images have some advantages over macroscopic photographs. They can be obtained quickly because the CT instrument can scan the entire body in several tens of seconds while it takes more time to acquire photographs and their quality depends on the photographer and the light conditions. Although advanced 3D surface imaging techniques using a multi-camera rig have been introduced to reduce the acquisition time, these systems are not yet widely used19.
With 3D CT imaging the forensic findings on axial images can be easily explained to non-radiologists5,6,7. Besides, the wound depth and the direction of the stab can be ascertained by confirming injuries and hemorrhages below the skin surface by axial CT images (see Fig. 6). Lastly, post-traumatic stress disorder for persons attending the trial is one of the most important issue in miserable case because they are forced directly to look at gruesome photographs20,21. The postmortem 3D CT image which is adjusted color tone may be effective to spare PTSD for them.
A 49-year-old woman killed with a knife. ((a) 3D CT GIR image, (b) axial CT image (soft kernel), (c) axial CT image (lung kernel)). (a) 3D CT GIR image showing a stab wound at the left anterior thorax (arrow). (b,c) The axial CT images show bilateral hemopneumothorax suggesting that the knife stabbed into the left anterior thorax penetrated the mediastinum and reached the right lung.
On the other hand, macroscopic photographs have a significant advantage, i.e. they provide the color information important for the accurate assessment of the wound edges. Measurements on the corpse and photographs providing color information are indispensable for the primary evaluation. Since CT images and macroscopic photographs each have advantages, we should use them according to a purpose for forensic investigations.
Measurement of the stab-wound diameter on 3D CT images
The mean difference between the diameter of stab wounds recorded at autopsy and on 3D CT VR and GIR images was 0.2 cm; GIR failed to lower the difference. The quality of 3D CT images is highly dependent on the quality of the original dataset7. To improve the quantitative assessment of 3D CT images, ultra-high resolution CT scanning may be required because its special resolution is superior to conventional CT22,23,24,25. In addition, in order to improve the quality of 3D CT images, the wound should not be masked when CT is scanned. The wounds on the back or side may be masked due to a body bag, CT bed and their arms. Keeping the stab wound on the back away from the CT bed with a sponge or cushion, or raising the arm, is effective in improving the image quality of 3D CT images. Placing additional external markers (radiopaque rings) may help to detect stab wounds on the back or side of the victim’s body26.
Limitations
Our study has some limitations. The readers’ judgement that GIR were superior to VR images was subjective because there was no objective assessment way for 3D CT images available. Also, the study population was small and we did not assess stab wounds delivered by devices other than knives. Differences in minor settings, including the placement of the virtual light, the opacity, and the type of workstation used may have slightly affected the quality of VR and GIR images.
