Abstract
Objectives:
Evaluation of performance and forensic relevance of a novel, photorealistic, 3D reconstruction method (cinematic rendering, (CR)) in comparison with conventional post-mortem CT (PMCT) and volume rendering (VR) technique for visualization of traumatic injuries.
Methods:
112 pathologies (fractures, soft tissue injuries and foreign bodies) from 33 human cadavers undergoing whole body PMCT after traumatic death were retrospectively analyzed. Pathologies were reconstructed with CR and VR techniques. Fractures were classified according to their dislocation. Images were evaluated according to their expressiveness and judicial relevance by two forensic pathologists using a five-level Likert-scale (1: high expressiveness, 5: low expressiveness). They decided whether CR reconstructions were suitable for judicial reviews. The detection rate of pathologies was determined by two radiologists.
Results:
CR was more expressive than VR for all three trauma categories (p < 0.01) and than conventional CT when used for fractures with dislocation (p < 0.001), injuries of the ventral body surface (p < 0.001), and demonstration of foreign bodies (p = 0.033). CR and VR became more expressive with a higher grade of fracture dislocation (p < 0.001). 20% of all pathologies in the CR and VR reconstructions were not detectable by radiologists.
Conclusion:
CR reconstructions are superior to VR regarding the expressiveness. For fractures with substantial dislocation, soft tissue injuries, and foreign bodies in situ, CR showed a significantly better expressiveness than conventional PMCT. CR and VR have significant limitations in cases of fractures with minor dislocations and covered soft tissue injuries.
Advances in knowledge:
CR is a helpful tool to present pathologies found in PMCT for judicial reviews.
Introduction
Post-mortem CT (PMCT) has become an important part of forensic pathology in the last years, often preceding and complementing autopsy. CT is a widely available technique offering non-invasive diagnosis and allowing for screening and documentation of important findings in virtual autopsies.1,2
A major challenge for forensic radiologists is to make diagnostic findings from PMCT understandable to non-radiologists such as police representatives, lawyers, judges, jury members, and forensic pathologists.3,4 Optimal visualization of findings may improve the understanding of pathologies and the correlation with the conventional autopsy findings. PMCT frequently even helps to detect additional pathologies when compared to conventional autopsy.1,5–8
The main findings of PMCT are often presented with 2D greyscale slice images or with multiplanar reformations (MPR). 3D post-processing imaging techniques such as VR or surface-shaded display (SSD) have been shown to be helpful and improve visualization.9 In clinical routine, 3D reconstructions are frequently used allowing an easy spatial allocation, for example in therapy planning or follow-up.10 In PMCT, the use of VR 3D reconstructions complementing the radiological report is a common practice.6,11,12
Recently, a novel software (“Cinematic Rendering,” Syngo.via, Siemens Healthineers, Forchheim, Germany) for improved visualization of 3D CT reconstructions was introduced. Cinematic rendering (CR) enables photorealistic reconstructions with natural shadowing, depth, and opacity by using a complex lighting algorithm.13 The software offers different rendering presets which maps Hounsfield units (HU) to color values and allowing for modification of the rendering results, for example by removing different kinds of tissues like muscles, soft tissue, or blood vessels.
Several studies have compared the performance of CR to volume rendering (VR) and declared CR reconstructions to be more impressive with better shape perception and more natural depth.14 Ampanozi et al demonstrated a better performance of 3D reconstructions with CR compared to conventional CT regarding understandability, cost-efficiency, and suitability for the use in courtrooms.12
However, a detailed analysis of the performance of CR in different grades of fracture dislocation, soft tissue injuries, and for foreign bodies presentation in comparison with VR reconstructions in PMCT has not yet been conducted. Therefore, the aim of this study was to evaluate the diagnostic performance of CR reconstructions in PMCT after traumatic death in different types of injuries, grades of fracture dislocation as well as its forensic impact.
Methods and materials
Patients
This retrospective single center study was approved by the local ethics committee. 43 human cadavers underwent whole body PMCT before conventional autopsy between February 2016 and March 2019. Seven patients were excluded because of non-traumatic death and three patients were excluded due to technical issues (one dataset could not be imported into the syngo.via software, two CT examinations were of severely reduced image quality due to soil around the corpse). Therefore, 33 human cadavers (8 female, 25 male) with a mean age of 54 ± 18 years (range 31–92 years) and a traumatic cause of death were included in our analysis (Figure 1). An exact calculation of the post-mortem interval (death to PMCT) cannot be given, as the time of death was not known for the investigated cases. Post-mortem interval ranged from half a day to three days maximum after admission of the corpse to the institute.
Figure 1.
Number of included cadavers.
PMCT scan protocol
21 PMCT scans were performed on a single source 64 slices CT scanner (Somatom Definition Edge, Siemens Healthineers, Forchheim, Germany) and 12 scans were performed on a dual source 2×64 slice CT scanner (Somatom Definition Flash, Siemens Healthineers, Forchheim, Germany).
As specified in our standard whole-body protocol for PMCT examinations, scans were performed with automated tube voltage selection (CarekV, Siemens Healthineers) with a reference value of 120kVp. Automated tube current modulation (CareDose 4D, Siemens Healthineers) was activated with a reference tube-current time product of 290 mAs. CT data were acquired with a 0.5 s rotation time, 128×0.6 mm collimation and a pitch of 0.55.
Arms were positioned alongside the torso to be included in the field of view whenever possible. All examinations were performed without contrast media.
Images were reconstructed using iterative reconstruction with a medium smooth kernel (I30f), a very sharp kernel (I70f), a slice thicknesses of 1 mm, an increment of 1 mm, and a matrix size of 512×512 pixels.
Reconstruction and classification of findings
All PMCT reports and autopsy reports were screened for pathologies with the radiological report as the gold standard. In the included 33 human cadavers, 112 major findings were detected, including 83 fractures, 24 soft tissue injuries, and five cases of foreign bodies like remaining knives (Figure 2). In order to evaluate the influence of fracture dislocation, fractures were classified according to their dislocation (Grade 1: no dislocation, Grade 2: minor dislocation, Grade 3: substantial dislocation).
Figure 2.
Number of reconstructed major findings, classification according to type of injury, fracture dislocation and location of soft tissue injury.
The 24 soft tissue injuries were divided into findings on the patient’s front site (n = 18) and findings on the patient’s backside or covered by other body parts (n = 6).
Image reconstruction
First, all main pathologies were displayed on conventional multiplanar images in axial, sagittal, and coronal multiplanar reconstructions using the institutional Picture Archiving and Communication System (PACS) (Sectra, V.19.3, Linköping, Sweden). Soft tissue injuries and foreign bodies were demonstrated in soft tissue window. Bone window was used for fractures.
The findings were reconstructed with the CR technique. For this purpose, CT reconstructions (1 mm slice thickness, I30f kernel) were transferred to a workstation with syngo.via software (Syngo.via, Siemens Healthineers) using the multi-modality reading workflow. Fractures were processed using three CR gallery settings adjusted to bones (“Osseous Tissue,” “Osseous and Soft Tissue,” “Default”). Soft tissue injuries were reconstructed using three CR gallery presets adjusted to soft tissue (“Patient marking,” “Dark Myocard,” “Default”). Foreign bodies were reconstructed using default software settings. Automatic table removal was applied.
Later, pathologies were carefully displayed using the zooming and rotation features. The cutting tool was used to display solely major findings which were saved as screenshots in each gallery setting. While courtroom demonstration and judicial reports do not allow using software, screenshots were taken in the three different gallery settings from a suitable angle for each major finding. Then, the subjectively most expressive reconstruction image was chosen from the different gallery setting screenshots by the radiologists in consensus.
Using the institutional PACS described above, additional VR reconstructions were made for each major finding. Automatic table removal was used if possible as well as the cutting tool. A screenshot of the VR reconstruction was taken from a similar angle as the CR reconstruction image. Gallery settings “Bone 1” and “Skin” were used for the VR reconstructions.
CR and VR post-processings of all pathologies were performed by the same resident radiologist with 3 years of experience in PMCT.
Subjective analysis by pathologists
Two forensic pathologists evaluated the selected 2DCT, CR, and VRT images from the screenshots independently according to their expressiveness of pathology on a five-point Likert-scale (1:high expressiveness, very good visualization of the finding; 2: good expressiveness, good visualization of the finding; 3: medium expressiveness, finding can be detected; 4: lower expressiveness; finding can hardly be detected; 5: very low expressiveness, unevaluable, finding cannot be detected).15
In addition, the forensic pathologists decided whether cinematic reconstructions are suitable for judicial reviews and demonstrations (1: useful; 0: not useful).
Subjective analysis by radiologists
The two radiologists (7 years and 3 years of experience in reading PMCT) independently analyzed the CR and VR reconstructions. The PMCT images were analyzed at least 2 months apart. For each pathology, the radiologists decided whether it was detectable in CT, CR, and VRT images, respectively, (1: detectable; 0: not detectable). The review for our study was performed at least 2 months after the initial written report of the PMCT to exclude a potential memory bias.
Statistical analysis
Statistical analysis was performed using SPSS V.26.0 (IBM Corp., Chicago, IL, USA). For the subjective analysis of expressiveness by forensic pathologists, median values were determined.
The results of the evaluation made by the forensic pathologists between the performance of CT and CR images and between the CR and the VR images were compared using the Wilcoxon singed-rank test. For the correlation between the grade of fracture dislocation and the evaluation by forensic pathologists, Spearman’s rank-order correlation was used with rs < 0 for negative correlation. A p-value of less than 0.05 indicated statistical significance.
Results
Subjective analysis of expressiveness by forensic pathologists
The median expressiveness value of all 112 CR reconstructions as well as conventional PMCT rated by forensic pathologists was 2.0 (range 1–5). All VR reconstructions were rated with a median value of 3.0 (range 1–5). Comparing the CR and VR reconstructions for all 112 major findings, the expressiveness of CR was significantly higher (p < 0.001). 57% of the 112 CR reconstructions were declared as suitable and useful for judicial reviews.
Looking at all fractures, we also found a median expressiveness of 2.0 in both CT and CR images with no significant difference in expressiveness. VR images were rated as medium expressive (median 3.0). Images of all fractures seen in total were significantly more expressive in CR than in VR reconstructions (p < 0.001). 58% of the CR reconstruction images were declared useful for judicial reports. Fractures with no dislocation were given a median expressiveness of 2.0 in CT images, of 5.0 in VR images, and of 4.0 in CR images reconstructions with only 13% of CR images declared useful for reports.
Fractures became significantly more expressive with a higher grade of dislocation in CR images (minor dislocation median 2.0 with 56% useful; substantial dislocation median 1.0 with 84% useful; p < 0.001) as well as in VR images (minor dislocation median 3.0), substantial dislocation median 2.0; p < 0.001) (Figures 3 and 4).
Figure 3.
Distribution of subjective evaluation for expressiveness by forensic pathologists on Likert-scale in different grades of fracture dislocation.
Figure 4.
Rib fractures with substantial dislocation of the third rib (green circles). Left to right: CT given a Likert Scale of 2 (good expressiveness, good visualization of the finding), CR given a Likert Scale of 1 (high expressiveness, very good visualization of the finding), VR with a Likert Scale of 2 (good expressiveness).
The CT images showed a median of 2.0 for every grade of dislocation with no significant correlation between the different grades (p = 0.678).
Soft tissue injuries were given a median expressiveness of 3.0 in CT, 2.0 in CR, and 3.5 in VR reconstructions with 46% of the CR images declared useful for judicial reviews. Findings on the patient’s front side (n = 18) were given a median value of 3.0 in CT, 2.0 in CR, and 3.0 in VR with 53% useful CR images, whereas findings on the patient’s backside or covered by the table or other body parts (n = 6) were rated with a median of 4.0 in CT, 4.0 in CR, and 5.0 in VR with only 25% useful CR images.
Reconstructions of soft tissue injuries on the front side of the corpse were significantly more expressive with CR reconstructions (median 2.0) than for CT (median 3.0; p < 0.001) and for VR reconstructions (median 3.0; p < 0.001). A significant correlation was found for the expressiveness and the localization of soft tissue injuries according to patient’s side in CR reconstructions (p < 0.001).
The superior expressiveness of CR images (median 1.0) concerning foreign bodies in comparison with CT (2.0, p = 0.033) and VR (2.0, p = 0.007) was also significant (Figure 5).
Figure 5.
Remaining knife. Left to right: CT image with a Likert Scale of 3 (medium expressiveness, finding can be detected), CR with a Likert Scale of 1 (high expressiveness, very good visualization of the finding), VR with a Likert Scale of 2 (good expressiveness).
Detection of pathologies by radiologists
Both radiologists detected all 112 pathologies (100%) in the CT images, whereas 22/112 pathologies (20%) were not detectable on CR and VR reconstructions alone. Pathologies missed on CR and VR reconstructions were fractures with no or slight dislocation (n = 17), orbital fractures (n = 1, see Figure 6) and soft tissue injuries covered by the CT table or other body parts (n = 4). Out of the 83 fractures, 65 (78%) were detectable by radiologists in CR reconstruction images. Out of the 37 fractures with substantial dislocation, 36 (97%) were detectable in CR reconstruction images. Foreign bodies showed a detection rate of 100% in CR reconstructions.
Figure 6.
Blowout fracture of left orbita (blue arrow). Left to right: CT reconstruction with a Likert Scale of 1 (high expressiveness, very good visualization of the finding), CR with a Likert Scale of 4 (lower expressiveness, findings can hardly be detected), VR with a Likert Scale of 5 (very low expressiveness, findings cannot be detected).
Regarding the detection rate of pathologies by radiologists, there was no difference between CR and VR reconstructions, but a significant difference between CR reconstructions and original PMCT images which were superior for all pathologies (p < 0.00001).
Discussion
PMCT with the possibility of post-processing imaging with CR and other reconstruction methods is a non-invasive technique of great value, also with regard to piety, cultural conditions, and lack of availability of conventional invasive autopsy.2,6 The use of 3D reconstructions as an alternative for 2D greyscale PMCT images for medical reports in courtrooms has already been described.12 Further studies showed that images created with CR are more photorealistic than those created with other established 3D reconstruction techniques like VRT.11 The superiority of previously available 3D reconstruction methods regarding understandability in courtrooms in comparison with greyscale CT images, VRT, or written reports only has also been demonstrated.11,12 The radiologists’ challenge is making PMCT findings comprehensible across professional groups, such as forensic pathologists, police, or justice members. The loss of information in the transmission process for radiological and autopsy findings due to a lack of comprehension or a pre-selection of single slices from PMCT has to be minimized. Especially regarding the lack of knowledge for human anatomy and an untrained spatial orientation sense, PMCT slices can be difficult to understand.
Interalia, the present study points out that the informative value and expressiveness of CR reconstructions to observers varies based on different parameters like the type of injury and dislocation of fractures. After further differentiation of major findings, a superior expressiveness for CR in comparison with CT images was found for fractures with substantial dislocation, foreign bodies in situ, and soft tissue injuries on the body surface not covered by the CT table or other body parts.
CT showed the same evaluation for expressiveness given by forensic pathologists in every grade of dislocation, whereas a strong correlation between grade of dislocation for fractures and CR expressiveness was found, as well as for VR reconstruction images. Reconstructions of fractures with major dislocation were declared useful for forensic reports in 84% of the cases. In cases of fractures of smaller bones like the orbita or midface, forensic pathologists often found the CR and VR reconstructions not useful (Figures 5 and 6). A possible explanation could be lower digital resolution for smaller bones embedded in soft tissue. In this case, forensic pathologists could also keep in mind the possibility of analyzing the bone through explicit surgical exposure during autopsy.
Capturing mainly the surface of structures, fractures without dislocation and slight affection of the corticalis are consequently missed in CR reconstructions. In these cases, conventional CT imaging was preferred.
The overall detection rate for all major pathologies found by radiologists was 100% in PMCT and 80% in CR and VR reconstructions. The detection rate for fractures was 100% for CT and 78% for CR and also VR reconstructions. Our study showed that fractures with no or slight dislocation and orbital fractures were missed on CR reconstructions as well as in VR images, whereas 97% of all fractures with substantial dislocation were detected in CR and VR reconstructions. The detection rate of fractures without dislocation was only 40%. Hence, compared to conventional CT images, CR and VR reconstructions alone are inferior and cannot meet the diagnostic standard.
Further studies already pointed out the great value of PMCT for revealing foreign bodies, especially to facilitate subsequent autopsy.16 In our cases, remaining knives, a projectile, osteosynthesis screws, and an intraosseous access were analyzed. Radiological analysis showed that all foreign bodies have been detected in CR reconstructions. Moreover, forensic analysis showed that CR reconstructions were more expressive than CT and VR images considering these findings. Therefore, orientation and localization of foreign bodies inside human corpses seem to be easier in CR reconstructions, when images where interpreted by non-radiologists (Figure 5).
PMCT has also been found to be a useful method for the visualization of stab wounds in other studies.17 In case of CR reconstruction and as well for VR reconstructions, we pointed out some limitations for the visualization of soft tissue injuries. 100% of all soft tissue injuries not covered by other body parts were detected by radiologists in our study, whereas only 33% of all injuries on the patients’ backside covered by the table or other body parts were detectable, even with the use of software tools like cutting out adjacent parts. A possible solution for a better detection of back lesions would be an additional procubitus PMCT scan of the body after turning it around, which was not performed in this study.
Therefore, we assumed that a good positioning, if necessary, with relocation of the body, leads to a better usability of the 3D reconstructions and a higher expressiveness than with conventional CT images.
Our study showed that CR reconstructions are helpful for forensic demonstrations, even although pathologists only found 57% of all 112 CR reconstructions suitable for courtroom demonstrations. Further, CR was superior to its predecessor VR technique regarding the expressiveness in every trauma category. Recently, Wollschlaeger et al published a study comparing the value of CR to conventional VRT images. Their findings showed a more detailed and helpful visualization of lower extremity fractures in CR reconstructions compared to conventional VR technique.18
Forensic pathologists are more familiar with CT than police or justice members who were not involved in this study, so looking at possible limitations, the value of CR might be underrated. In addition, PMCT was performed without contrast media leading to a known lower sensitivity.
Only images of screenshots were used for the evaluation in our study since there is no possibility of using special software in courtrooms and judicial reports in our region. Hence, the great advantages of features like rotation or zooming in the whole examination used in software could not be used for demonstration and were neglected in this study. Furthermore, injuries to internal organs like liver ruptures or collapsed lungs were not taken into account in this study.
Conclusion
CR reconstructions are superior to VR reconstructions regarding the expressiveness in all three trauma categories. CR was more useful than conventional PMCT slices in cases of fractures with substantial dislocation, demonstration of foreign bodies, and body surface injuries not covered by other body parts or the CT table. Especially regarding fractures without substantial dislocation and covered soft tissue injuries, CR cannot be used for diagnostic purposes without PMCT. Taking the limitations into account, CR can improve the interdisciplinary information transfer and the comprehension of forensic findings.
Contributor Information
Judith Böven, Email: judith.boeven@med.uni-duesseldorf.de.
Johannes Boos, Email: johannesboos@gmx.de.
Andrea Steuwe, Email: andrea.steuwe@med.uni-duesseldorf.de.
Janna Morawitz, Email: Janna.morawitz@med.uni-duesseldorf.de.
Lino Morris Sawicki, Email: LinoMorris.Sawicki@med.uni-duesseldorf.de.
Julian Caspers, Email: Julian.Caspers@med.uni-duesseldorf.de.
Lisa Küppers, Email: lisa.kueppers@med.uni-duesseldorf.de.
Benno Hartung, Email: benno.hartung@med.uni-duesseldorf.de.
Christoph Thomas, Email: thomas@radiologiekrefeld.de.
Gerald Antoch, Email: antoch@med.uni-duesseldorf.de.
Joel Aissa, Email: joel.aissa@med.uni-duesseldorf.de.
REFERENCES
- 1.Roberts ISD, Benamore RE, Benbow EW, Lee SH, Harris JN, Jackson A, et al. Post-Mortem imaging as an alternative to autopsy in the diagnosis of adult deaths: a validation study. The Lancet 2012; 379: 136–42. doi: 10.1016/S0140-6736(11)61483-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Blum A, Kolopp M, Teixeira PG, Stroud T, Noirtin P, Coudane H, et al. Synergistic role of newer techniques for forensic and postmortem CT examinations. AJR Am J Roentgenol 2018; 211: 3–10. doi: 10.2214/AJR.17.19046 [DOI] [PubMed] [Google Scholar]
- 3.Ampanozi G, Zimmermann D, Hatch GM, Ruder TD, Ross S, Flach PM, et al. Format preferences of district attorneys for post-mortem medical imaging reports: Understandability, cost effectiveness, and suitability for the courtroom: a questionnaire based study. Leg Med 2012; 14: 116–20. doi: 10.1016/j.legalmed.2011.12.008 [DOI] [PubMed] [Google Scholar]
- 4.Araki A, Ishikawa N, Takami S, Ishikawa N, Amano C, Takeshita H, et al. Interpretation of postmortem head computed tomography for non-traumatic in-hospital deaths by non-radiologists: a preliminary study. Springerplus 2016; 5: 978. doi: 10.1186/s40064-016-2653-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Grabherr S, Heinemann A, Vogel H, Rutty G, Morgan B, Woźniak K, et al. Postmortem CT angiography compared with autopsy: a forensic multicenter study. Radiology 2018; 288: 270–6. doi: 10.1148/radiol.2018170559 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Norberti N, Tonelli P, Giaconi C, Nardi C, Focardi M, Nesi G, et al. State of the art in post-mortem computed tomography: a review of current literature. Virchows Arch 2019; 475: 139–50. doi: 10.1007/s00428-019-02562-4 [DOI] [PubMed] [Google Scholar]
- 7.Del Fante Z, De Matteis A, Fazio V, Di Fazio N, Quattrocchi A, Romano S, et al. The importance of post mortem computed tomography (PMCT) in the reconstruction of the bullet trajectory. Clin Ter 2019; 170: e90. doi: 10.7417/CT.2019.2122 [DOI] [PubMed] [Google Scholar]
- 8.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. J Forensic Sci 2003; 48: 2002166–403. doi: 10.1520/JFS2002166 [DOI] [PubMed] [Google Scholar]
- 9.Calhoun PS, Kuszyk BS, Heath DG, Carley JC, Fishman EK. Three-Dimensional volume rendering of spiral CT data: theory and method. Radiographics 1999; 19: 745–64. doi: 10.1148/radiographics.19.3.g99ma14745 [DOI] [PubMed] [Google Scholar]
- 10.Coll DM, Herts BR, Davros WJ, Uzzo RG, Novick AC. Preoperative use of 3D volume rendering to demonstrate renal tumors and renal anatomy, 20 (n.d.. 8. [DOI] [PubMed] [Google Scholar]
- 11.Ebert LC, Schweitzer W, Gascho D, Ruder TD, Flach PM, Thali MJ, et al. Forensic 3D visualization of CT data using Cinematic volume rendering: a preliminary study. AJR Am J Roentgenol 2017; 208: 233–40. doi: 10.2214/AJR.16.16499 [DOI] [PubMed] [Google Scholar]
- 12.Ampanozi G, Zimmermann D, Hatch GM, Ruder TD, Ross S, Flach PM, et al. Format preferences of district attorneys for post-mortem medical imaging reports: understandability, cost effectiveness, and suitability for the courtroom: a questionnaire based study. Leg Med 2012; 14: 116–20. doi: 10.1016/j.legalmed.2011.12.008 [DOI] [PubMed] [Google Scholar]
- 13.GmbH SH. syngo.via Cinematic VRT white paper. Cinematic Rendering for Medical Imaging 2017;. accessed July 28, 2019. [Google Scholar]
- 14.Dappa E, Higashigaito K, Fornaro J, Leschka S, Wildermuth S, Alkadhi H. Cinematic rendering - an alternative to volume rendering for 3D computed tomography imaging. Insights Imaging 2016; 7: 849–56. doi: 10.1007/s13244-016-0518-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Jamieson S. Likert scales: how to (ab)use them. Med Educ 2004; 38: 1217–8. doi: 10.1111/j.1365-2929.2004.02012.x [DOI] [PubMed] [Google Scholar]
- 16.Sano R, Takahashi Y, Hayakawa A, Murayama M, Kubo R, Hirasawa S, et al. Use of postmortem computed tomography to retrieve small metal fragments derived from a weapon in the bodies of victims in two homicide cases. Leg Med 2018; 32: 87–9. doi: 10.1016/j.legalmed.2018.03.006 [DOI] [PubMed] [Google Scholar]
- 17.Schnider J, Thali MJ, Ross S, Oesterhelweg L, Spendlove D, Bolliger SA. Injuries due to sharp trauma detected by post-mortem multislice computed tomography (MSCT): a feasibility study. Leg Med 2009; 11: 4–9. doi: 10.1016/j.legalmed.2008.07.001 [DOI] [PubMed] [Google Scholar]
- 18.Wollschlaeger LM, Boos J, Jungbluth P, Grassmann J-P, Schleich C, Latz D, et al. Is CT-based cinematic rendering superior to volume rendering technique in the preoperative evaluation of multifragmentary intraarticular lower extremity fractures? Eur J Radiol 2020; 126: 108911. doi: 10.1016/j.ejrad.2020.108911 [DOI] [PubMed] [Google Scholar]