www.elsevier.com/locate/legalmed

Legal Medicine 9 (2007) 100–104

VIRTOPSY – The Swiss virtual autopsy approach

Michael J. Thali *, Christian Jackowski, Lars Oesterhelweg,Steffen G. Ross, Richard Dirnhofer

Institute of Forensic Medicine, Center of Forensic Imaging/Virtopsy, University of Berne, Switzerland

Abstract

The aim of the VIRTOPSY project (www.virtopsy.com) is utilizing radiological scanning to push low-tech documentation and autop-sy procedures in a world of high-tech medicine in order to improve scientific value, to increase significance and quality in the forensicfield. The term VIRTOPSY was created from the terms virtual and autopsy: Virtual is derived from the Latin word ‘virtus’, which means‘useful, efficient and good’. Autopsy is a combination of the old Greek terms ‘autos’ (=self) and ‘opsomei’ (=I will see). Thus autopsymeans ‘to see with ones own eyes’. Because our goal was to eliminate the subjectivity of ‘‘autos’’, we merged the two terms virtual andautopsy – deleting ‘‘autos’’ – to create VIRTOPSY. Today the project VIRTOPSY combining the research topics under one scientificumbrella, is characterized by a trans-disciplinary research approach that combines Forensic Medicine, Pathology, Radiology, Image Pro-cessing, Physics, and Biomechanics to an international scientific network. The paper will give an overview of the Virtopsy change processin forensic medicine.� 2006 Elsevier Ireland Ltd. All rights reserved.

Keywords: Forensic radiology; Virtopsy; Virtual autopsy; Autopsy imaging

1. Introduction

The application of imaging methods for non-invasivedocumentation and analysis of relevant forensic findingsin living and dead persons has lagged behind the enormoustechnical development of imaging methods. There are onlya few textbooks dealing with forensic radiology [1,2]. Mostof these textbooks concentrate on classical roentgeno-graphic methods and hardly cover the newer sectionalimaging techniques of computed tomography and magnet-ic resonance imaging in detail. Forensic radiology, includ-ing all techniques and their many uses for forensicpurposes, now is a rapidly growing interdisciplinary sub-specialty of both forensic medicine and radiology. Shortlyafter the communication of the detection of X-rays byConrad Roentgen the new non-invasive technique was usedfor forensic documentation purposes. But modern cross-

1344-6223/$ - see front matter � 2006 Elsevier Ireland Ltd. All rights reserve

doi:10.1016/j.legalmed.2006.11.011

* Corresponding author. Tel.: +41 31 631 84 12; fax: +41 31 631 38 33.E-mail address: michael.thali@irm.unibe.ch (M.J. Thali).URL: www.virtopsy.com (M.J. Thali).

section imaging is still underutilized in forensics, mainlydue to the unawareness of its potential in forensic sciencebut also to the cost and the limited access to and trainingfor these newer modalities, such as Computer Tomogra-phy-CT, including spiral multislice, and MagneticResonance Imaging-MRI.

2. The Swiss virtual autopsy project (VIRTOPSY)

2.1. Materials and methods

The Institutes of Forensic Medicine and of DiagnosticRadiology of the University of Bern, Switzerland, starteda research project in 2000, with the hypothesis that non-in-vasive imaging might predict autopsy findings and maybegive additional information.

The responsible justice department and also the ethicscommittee of the University of Bern approved the study.

In this joint project called ‘‘Virtopsy’’ [11] we used thenewest generation of:

d.

Fig. 1. Forensic CT examination of a dead body at the University of Bern.

Fig. 2. 3D-MSCT of an gunshot injury. The exit wound is visible bylooking through the entrance wound.

Fig. 3. Gunshot injury to the head: MRI is showing bullet wound track inthe cerebellum (arrow).

M.J. Thali et al. / Legal Medicine 9 (2007) 100–104 101

• Multidetector row or Multislice Spiral ComputerTomography (MSCT),

• and 1.5 Tesla MR scanner from GE with Spectroscopysoftware.

In special situations bone tissue specimens were examinedon a Micro-CT and Micro-MR system. The Micro-CT isdeveloped and built at the Institute of Medical Physics Erlan-gen, Germany. This Micro-CT scanner can image a 3D vol-ume with an isotropic resolution for resolution ranges from10 to 100 lm. The system is capable of examining sampleswith diameters from 4 to 40 mm. The Micro-MR studieswere performed on a Bruker DMX spectrometer (BrukerBiospin MRI, Inc., Billerica, MA) coupled to a wide-boremagnet operating at 9.4 T (400 MHz for protons).

At the beginning in 2000, all dead bodies were transport-ed by undertakers to the hospital for the CT and MRIscanning. Wrapping the corpses in artifact-free body bags,as requested by the ethics committee, preserved anonymityof the deceased. Since 2005 the Forensic Institute at theUniversity of Bern has its own Siemens MSCT scanner(Siemens� high speed type 6 thin-slices at every rotation)for the postmortem scanning.

By now, 100 forensic cases have received a full bodyexamination by CT and MRI before autopsy (Fig. 1).

The results of CT and MRI were correlated with thefindings of autopsy [3,4,7–10], analyzing the indicationsof each type of exploration depending on the expectedpathology.

2.2. Results

The scan times are short from 1 to 10 min.The correlation with the forensic autopsy findings

showed:

2.2.1. Computer Tomography-CT

The CT depending on the slice thickness and the volumeto be covered, have been found a superior tool for 2D and3D documentation and analysis of

• fracture systems,• pathologic gas collections (whether air embolism, subcu-

taneous emphysema after trauma, hyperbaric trauma, ordecomposition effects),

• and it also shows gross tissue injury.

Post-processing on commercial scanning workstationwith 3D SSD (Surface Shade Display) and VR (VolumeRendering) can provide useful visualization for court trial(Fig. 2).

For example, in gunshot cases the determination ofentrance and exit wounds is possible based on the charac-teristic fracture pattern with inward or outward bevelingof the bone respectively.

CT and MRI are excellent tools to visualize bullet tractswith hemorrhage (Figs. 2 and 3). Metal artifacts due to the

102 M.J. Thali et al. / Legal Medicine 9 (2007) 100–104

bullet can appear on CT images; these effects will be reducedin the near future by metal artifact reduction algorithms.

As compared to clinical imaging in trauma or forensicvictims, the major drawback of postmortem CT is the lack-ing availability of intravenous contrast enhancement aftercirculatory arrest, which makes analysis of parenchymaand vascular injury much more difficult, less sensitive andless specific.

2.2.2. Magnetic Resonance-MR

In demonstrating soft tissue injury, neurological andnon-neurological organ trauma, and non-traumatic

Fig. 4. (a) Autopsy image: knife wound injury to the heart (arrow). (b)Corresponding finding in MRI: knife wound to the heart (arrow).

pathology, the MRI (Figs. 4 and 5), compared to CT,clearly had a

• higher sensitivity,• higher specificity, and• higher accuracy

Studies of child abuse victims confirm the sensitivity ofpostmortem MRI for contusion, shearing injuries andsubdural hematoma.

Differences in morphology and signal characteristicsbetween antemortem and postmortem MRI do exist; howev-er, they have not yet been studied systematically.

If the results of clinical MRI can be transferred to post-mortem analysis, there is a great future for non-destructiveanalysis of visceral pathology, such as cardiac (includingcoronary), pulmonary and hepatic disease.

2.2.3. Magnetic Resonance Spectroscopy-MRS

Finally, MRS, combined with MRI, has a great poten-tial in documenting pre-terminal and postmortem metabo-lite concentrations in tissues. Since decompositioncontinuously changes the concentration of chemical com-pounds postmortem, MRS might be helpful in determiningthe time of death [13–15].

3. Forensic application of radiological micro-imaging –

Virtual histology

In many cases, the resolution of clinical scanners is notsufficient to answer questions relevant to forensic medicinenondestructively. This favors the idea of using microscop-ing non invasive imaging methods with their much higherresolution to visualize forensic specimens [4].

We have used microtomography of small object ormicro-CT in a forensic case of a knife’s blade inside corticaland trabecular bone to determine the injury pattern and theweapon involved [4,12].

In forensic soft tissue injury, retinal hemorrhage andelectric injury to the skin were studied by micro-MR(MR microscopy) [5,6].

We expect these new radiological cross-sectionalmicro-imaging methods to have a comparable impact on(forensic) histopathology, leading to virtual histology.

4. Data management and teleconsultation

The Virtopsy project generates enormeous numbers ofdigital DICOM data that can easily be archived, transmit-ted on a network, copied, quantitatively analyzed and post-processed on a workstation.

Digital format not only allows compact digital archivingbut also cuts the cost of films, of film handling and ofarchive space as soon as an institute is prepared forthe digital solution (PACS = picture archiving andcommunication system).

Fig. 5. (a) Autopsy image: rupture of aorta (arrow). (b) Correspondingfinding in MRI: aortic rupture (arrow).

M.J. Thali et al. / Legal Medicine 9 (2007) 100–104 103

Postprocessing is another tool that opens new ways ofanalyzing imaging data. Image contrast can be enhanced,distances, areas and volumes measured, and advanced soft-ware programs will help the doctor find tiny pathologicfindings.

Finally, teleradiology will open new teleconsulting ser-vices in the near future. In Switzerland, the aspects of tele-consulting of such forensic data is under discussion Nodoubt, forensic radiology will similarly share the advancesof clinical imaging.

5. Conclusion and outlook

Evidently, imaging techniques are nowadays excellenttools for forensic medicine. Similar to inspection and pho-tography but in contrast to other tools, they are able tofreeze the findings at the moment of investigation withoutcausing any damage. Freezing means permanent (analogueor digital) preservation as a document of proof, whetherthe victim is dead and undergoing postmortem decay orsurviving and loosing evidence due to healing. Causingno damage is an essential prerequisite in a living personthat is fulfilled indisputably. Even in dead persons, non-destructive documentation is important for two reasons:

1. First, it brings its information without precluding anyother conservative or destructive forensic investigation.

2. Second, it can be used in cultures and situations whereautopsy is not tolerated by religion or rejected by familymembers.

Whether and to what degree radiological minimallyinvasive ‘‘virtual autopsy’’ will in defined situations replacethe classical dissection technique will be decided in the nearfuture.

Two innovative forensic documentation methods arerising at the horizon:

1. the combination of sectional imaging with surface docu-mentation methods, such as photogrammetry and 3Doptical scanning, and

2. the combination of noninvasive imaging with minimallyinvasive image-guided tissue sampling from any bodylocation needed [7–10]. Tissue samples can be used forcytology, histology, chemical, and microbiologicalanalysis.

Radiologic virtual autopsy offers other advantages, suchas

1. an easy examination of bodies contaminated by infec-tion, toxic substances, radionuclides or otherbiohazards.

2. 2D and 3D postprocessing incredibly helps to visualizethe findings to people not present during the examina-tion, e.g., in court.

3. Complete, easily retrievable digital archives and telecon-sultation will support the process of qualityimprovement.

To support this process we founded the Technical Work-ing Group Forensic Imaging Methods (www.twgfim.com).Forensic Imaging will be an exciting science in the future.

Acknowledgements

Thanks go to all the Virtopsy research team members(see [11]).

104 M.J. Thali et al. / Legal Medicine 9 (2007) 100–104

References

[1] Brogdon BG. Forensic Radiology. Boca Raton: CRC Press;1998.

[2] Hart BL, Dudley MH, Zumwalt RE. Postmortem cranial MRI andautopsy correlation in suspected child abuse. Am J Forensic MedPathol 1996;17(3):217–24.

[3] Thali MJ, Yen K, Schweitzer W, Vock P, Boesch C, Ozdoba C, et al.Virtopsy, a new imaging horizon in forensic pathology: virtualautopsy by postmortem multislice computed tomography (MSCT)and magnetic resonance imaging (MRI) – a feasibility study. JForensic Sci 2003;48(2):386–403.

[4] Thali MJ, Taubenreuther U, Karolczak M, Braun M, BrueschweilerW, Kalender WA, et al. Forensic microradiology: micro-computedtomography (Micro-CT) and analysis of patterned injuries inside ofbone. J Forensic Sci 2003;48(6):1336–42.

[5] Thali MJ, Dirnhofer R, Becker R, Oliver W, Potter K. Is ‘virtualhistology’ the next step after the ’virtual autopsy’? Magneticresonance microscopy in forensic medicine. Magn Reson Imaging2004;22(8):1131–8.

[6] Thali M, Potter K, Dirnhofer R. From Virtopsy to Micro-Virtopsy:Virtual Forensic Histology. 81th Annual Congress of the GermanForensic Society, in Rostock, Germany, 2003.

[7] Thali M, Braun M, Kneubuehl B, Brueschweiler W, Vock P,Dirnhofer R. Improved vision in forensic documentation: Forensic,3D/CAD-supported photogrammetry of bodily injury external sur-faces, combined with volumetric radiologic scaninng of bodily injuryinternal structures to provide more leads and stronger forensicevidence. Oliver W. 3D visualisation for data exploration anddecision making. SPIE 2000:213–21.

[8] Thali M, Braun M, Dirnhofer R. Optical 3D surface digitizing inforensic medicine: 3D documentation of skin and bone injuries.Forensic Sci Int 2003;48(6):1356–65.

[9] Thali M, Braun M, Wirth J, Vock P, Dirnhofer R. 3D Surface and3D body documentation in forensic medicine: 3D/CAD photogram-

metry merged with 3D radiological scanning. J Forensic Sci2003;48(6):1356–65.

[10] Thali MJ, Braun M, Buck U, Aghayev E, Jackowski C, Vock P, et al.VIRTOPSY – scientific documentation, reconstruction and anima-tion in forensic: individual and real 3D data based geometricapproach including optical body/object surface and radiologicalCT/MRI scanning. J Forensic Sci 2005;50(2):428–42.

[11] www.virtopsy.com.[12] Microphotonics. http://www.microphotonics.com/skymto.html.[13] Ith M, Bigler P, Scheurer E, Kreis R, Hofmann L, Dirnhofer R,

et al. Observation and identification of metabolites emergingduring postmortem decomposition of brain tissue by means ofin situ 1H-magnetic resonance spectroscopy. Magn Reson Med2002;48(5):915–20.

[14] Scheurer E, Ith M, Dietrich D, Kreis R, Husler J, Dirnhofer R, et al.Statistical evaluation of time-dependent metabolite concentrations:estimation of post-mortem intervals based on in situ 1H-MRS of thebrain. NMR Biomed 2005;18(3):163–72.

[15] Delnomdedieu M, Hedlund LW, Johnson GA, Maronpot RR.Magnetic resonance microscopy – A new tool for the toxicologicpathologist. Toxicol Pathol 1996;24:36–44

. Prof. Dr. med. Michael Thali, Executive MBAHSG, is working since 1995 in forensic medicine.He has a two year fellowship in clinical radiology.In 2001/2002 he was a fellow at the Armed ForcesInstitute of Pathology (AFIP) in Washington DC.He wrote many virtual autopsy papers (seewww.virtopsy.com). Since February 2006 he is fullprofessor for forensic medicine at the Universityof Bern, Switzerland. He is director of the ‘‘Centerfor Forensic Imaging’’ at the Institute of ForensicMedicine Bern.