Beruflich Dokumente
Kultur Dokumente
Abstract
After the death phenomenon, the rigor mortis development, characterized by body stiffening, is one of the most evident
changes that occur in the body. In this work, the development of rigor mortis was assessed using a skinfold caliper in human
cadavers and in live people to measure the deformation in the biceps brachii muscle in response to the force applied by the
device. Additionally, to simulate the measurements with the finite element method, a two-dimensional model of an arm section
was used. As a result of the experimental procedure, a decrease in deformation with increasing postmortem time was
observed, which corresponds to an increase in rigidity. As expected, the deformations for the live subjects were higher. The
finite element method analysis showed a correlation between the c1 parameter of the neo-Hookean model in the 4- to 8-h
postmortem interval. This was accomplished by adjusting the c1 material parameter in order to simulate the measured experi-
mental displacement. Despite being a preliminary study, the obtained results show that combining the proposed experimental
procedure with a numerical technique can be very useful in the study of the postmortem mechanical modifications of human
tissues. Moreover, the use of data from living subjects allows us to estimate the time of death paving the way to establish this
process as an alternative to the existing techniques. This solution constitutes a portable, non-invasive method of estimating the
postmortem interval with direct quantitative measurements using a skinfold caliper. The tools and methods described can be
used to investigate the subject and to gain epidemiologic knowledge on rigor mortis phenomenon.
Keywords
Biomechanical testing, finite element method, rigor mortis, postmortem interval, necromechanics
average of both measurements. With these thickness the medial cross-section of the arm, which corresponds
values, the displacement at a certain hour postmortem to the area where the measurements were performed,
was calculated by the following expression: was chosen. Only the left arm was used for the simula-
tions. The real dimensions of the arm were obtained
Dd = lt l0 through digital image analysis with the ImageJ soft-
with l0 being the initial thickness and lt the thickness ware, using the known dimension of the threaded holes
while the force was being applied, measured at a certain made to serve several purposes, such as to help in the
postmortem hour t. A decrease in the displacement alignment of the cuts.17
value with time indicates an increase in rigidity. The selected slice was manually segmented and the
To study the evolution of rigor mortis, the measure- contours of the different tissues that constitute the arm,
ments were performed every hour during 4 h (in some namely, the skin, fat, bone and the different muscles
cases during 5 h) starting from the first measurement. (biceps brachii, brachialis, and medial, long and lateral
The population of this study consisted of a total of 18 head of triceps brachii), were obtained. Afterwards, the
corpses, 6 females and 12 males, with an average age of contours were imported into the software ABAQUS
62.9 6 16.8 years, an average body mass index (BMI) of and the finite element mesh was created. The finite ele-
25.9 6 6.9 kg/m2 and a postmortem interval between 2 ment mesh consisted of 4281 four-node bilinear plane
and 13 Hpm. The exclusion parameters included death strain quadrilateral, hybrid, constant pressure elements
by drowning or fire, corpses with signs of cadaveric (CPE4H in ABAQUS). The clamps of the Adipsmeter
decomposition or with significant muscular atrophy were modeled as 2D analytic rigid shells, while con-
and corpses stored in the cold chamber prior to the straining the clamps’ rotations and their vertical move-
measurement. ment. The bone was considered as a rigid body. The
contact between the clamps and the skin was consid-
ered frictionless since the use of friction did not change
Measurements on live people the simulation results significantly. The simulation was
performed with an imposed load of magnitude equal to
To compare the values obtained in the corpses with the
the load applied by the device, 8.4 N. The final finite
strain value measured in live people, the measurements
element model can be observed in Figure 2.
were also performed in 58 volunteers, 31 females and
The material behavior for the different layers was
27 males, with an average age of 40.8 6 15.1 years and
assumed to be homogeneous, isotropic, hyperelastic
with an average BMI of 25.3 6 4.3 kg/m2. The proce-
and incompressible. Due to the presence of fibers, the
dure to obtain and analyze the data was identical to the
muscle behavior is normally assumed as transversely
one described previously. With these measurements,
isotropic, when a three-dimensional (3D) model is con-
the mean displacement value in the biceps brachii for
sidered. Since a plane strain condition is assumed, and
live people when subjected to the force applied by the
the muscle fiber orientation is perpendicular to the sec-
device was obtained. Additionally, this mean value also
tion considered in this study, as shown in Figure 2, the
gives an indication of the displacement expected to be
effect of the fibers was neglected and the muscles
measured at the moment of death, that is, a baseline
assumed as isotropic. To characterize the mechanical
deformation.
behavior of the muscle and adipose tissue layers, the
Both live volunteers and corpses were from the
neo-Hookean constitutive model was chosen following
North region of Portugal. There was no rigid control
the approach used by Tran et al.18 to estimate the
of race; nevertheless, both groups were highly domi-
material constants for the different layers based on
nated by Caucasians, as is typical of the demographics
indentation tests. In the neo-Hookean model, the
of the region.
strain-energy density function for isotropic and incom-
pressible materials19 is given by (1)
Finite element analysis
C = c 1 ð I 1 3Þ ð1Þ
In this work, the compression of the biceps brachii in
response to the force applied by the Adipsmeter was where I1 is the first scalar invariant of the right
analyzed and computationally simulated using the Cauchy–Green tensor and is given by I1 = tr(C). In this
finite element method (FEM) with a simplified model work, the adipose tissue was modeled with a c1 material
in ABAQUSÒ. For that purpose, a 2D model of the constant of 0.42 kPa. By comparing the numerical
arm was used to analyze the behavior of the tissues results with the displacements observed experimentally
when subjected to an applied load. The measurements for live people and for the interval of 4–8 Hpm, the
performed allowed us to estimate the mechanical prop- muscle tissue material constant c1 of the neo-Hookean
erties of the different layers. The geometry of the arm model was manually adjusted in order to obtain a good
was reconstructed from the images obtained from the agreement between the FEM simulation and the experi-
Visible Human Project. The cadaver used in this proj- mental results.
ect was a 38-year-old male with a BMI of 27.9 kg/m2. Regarding the skin layer, different properties
For this work, a slice of the thorax region containing obtained from the literature were tested and the one
Figure 2. Finite element model used. The different materials are presented in different colors: skin in green, adipose tissue in
white, muscles in red and bone in blue. The two clamps of the device are presented in black at the initial position of the simulation.
presenting more suitable results regarding the experi- negative slope of approximately 20.50) with increasing
mental data was chosen. In that sense, the skin was Hpm, which points to an increase in the rigidity. Data
modeled using the properties from Evans and Holt,19 for 3 and 9 Hpm were not included in the trend line cal-
which uses the Ogden material model to characterize culations, despite fitting the linear model reasonably
the mechanical properties of the skin. In this model, well, since sample size was approximately 50% of the 4-
the strain-energy function19 is expressed in terms of the to 8-Hpm interval. Data for 2 Hpm and in the interval
principal stretches (2) from 10 to 13 Hpm had to be discarded since the sam-
ple size was too small due to logistic constraints. In
X
N
mp a a a order to compare the displacement measured in the
Cðl1 , l2 , l3 Þ = l1 p + l2 p + l3 p 3 ð2Þ
p=1
ap corpses and in live people, the measurements were also
performed in 58 volunteers. A higher average displace-
The versatility of Ogden’s model lies on its extensibility ment value was measured in the live people
as additional terms, corresponding to additional para- (31.27 6 4.62 mm) in comparison to the values mea-
meter pairs ðmi , ai Þ : mi ai . 0, may be added depending sured in the cadavers, as expected. Additionally, a com-
on problem complexity. It is widely used in many parison between the values measured on the left and
engineering modeling applications of hyperelastic right sides of the same muscular group was performed.
materials. In this work, the skin layer was modeled As a result, no considerable difference was found in the
using m = 10 Pa and a = 26. evolution of rigor mortis on both sides of the body for
most of the measured cadavers. Nevertheless, the
obtained values differed in the muscular group side in
Results the majority of the population both in the corpses and
Adipsmeter measurements in the live people. A difference between both sides of the
body is not surprising since usually people do not have
At each hour of the postmortem interval considered
equally developed muscles on both sides of the body.
(2–13 Hpm), both left arm and right arm measurements
were taken. Using the data obtained from the 18 cada-
vers, an average of the displacements at each Hpm was
Finite element analysis
calculated. This method allowed correlating the experi- In order to correlate the modification of the mechanical
mental values with the postmortem time (Figure 3). properties with the postmortem time, the material para-
For Hpm with the highest number of cases meter c1 of the neo-Hookean model for the muscle
(4–8 Hpm), a trend line was drawn using a linear regres- layer was manually adjusted so that the simulation out-
sion of the type, displacement = m 3 Hpm + b. The comes corresponded to the total displacement observed
results show that the displacement diminishes (with a experimentally in the live people and in the corpses for
Figure 3. Graphical representation of the average displacement values (mm) obtained in the biceps brachii muscle for the live
people and for the corpses according to the postmortem time. The vertical black lines represent the standard deviation. The gray
points represent the postmortem hours with the highest number of cases; for these points, a linear trend line was drawn and the
respective equation is shown in the top right corner.
Figure 5. Reconstructed geometry after deformation resultant of an 8.4-N load applied on each clamp: (a) deformed geometry and
(b) maximum principal stresses distribution (MPa).
establishment of a relation between the measured val- This general approach does not account for individual
ues and a physical parameter that allows the characteri- variability but is a prototype of the technique to be
zation of the tissue mechanical behavior is of great used with every subject in future tests.
importance. This parameter might be the Young modu- The obtained results allow us to have a first insight
lus for elastic materials, or in the case of hyperelastic into how the mechanical properties of the muscular tis-
materials, it can be the parameters of the constitutive sue change with the postmortem time. Considering that
models. In this work, a first correlation between the when working with human cadavers several restrictions
muscle material parameter c1 of the neo-Hookean and ethical protocols need to be followed, the use of
model and the time since death was already achieved. minimal invasive procedures is of great importance
In order to take more assertive conclusions, the sample when dealing with human corpses. For that reason, the
size should be increased, since this study was carried proposed experimental procedure combined with a
out with a population of 18 cadavers. This would allow numerical technique can be very useful in the study of
obtaining a more accurate average displacement value the mechanical modifications of human tissues after
for each hour postmortem. death. This solution constitutes a portable, non-
In addition, the method could be improved by the invasive method of estimating the postmortem interval
implementation of an inverse FEM allowing the esti- with direct quantitative measurements using a skinfold
mation of the mechanical properties from the measured caliper. The tools and methods described can be used
experimental displacement in a more expeditious way. to investigate the subject and to gain epidemiologic
Constructing a specific 2D geometry for each cadaver, knowledge on rigor mortis phenomenon. Moreover,
for example by the estimation of the thickness of each the use of data from living subjects allows us to esti-
layer using ultrasound imaging, would result in more mate the time of death as shown in Figure 4. If the
accurate simulations. Therefore, the lack of support technique is able to provide an adequate degree of
imaging techniques (X-ray, computed tomography accuracy in larger validation tests, it may be considered
(CT) or ultrasound) is a limitation of this study. as an alternative to estimate the time of death.
Afterwards, applying in the different geometries, the
displacement obtained during the experimental mea- Declaration of conflicting interests
surements and using an optimization algorithm, the
material parameters for each layer could be estimated None declared.
for different postmortem hours, and not only for the
muscle layer, as in this work. For that purpose, the Funding
contribution of each layer to the final displacement xão para
This study was financially supported by Fundac
value should be studied. To mitigate this limitation, the a Ciência e a Tecnologia (FCT)—Portugal, through the
2D finite element model built on one level of the visual Fundo Europeu de Desenvolvimento Regional
human project was chosen taking into account the typi- (FEDER) under Grant SFRH/BPD/71080/2010 from
cal biceps size found on both cadavers and volunteers. Fundo Social Europeu (FSE) under programs POPH-
QREN, and Project ‘‘Biomechanics: contributions to 10. Brand RA. Biomechanics or necromechanics? Or how to
the healthcare’’, reference NORTE-07-0124-FEDER- interpret biomechanical studies. Iowa Orthop J 2002; 22:
000035 co-financed by Programa Operacional Regional 110.
do Norte (ON.2—O Novo Norte), through the Fundo 11. Martins PALS, Jorge RMN and Ferreira AJM. A com-
Europeu de Desenvolvimento Regional (FEDER). parative study of several material models for prediction
of hyperelastic properties: application to silicone-rubber
and soft tissues. Strain 2006; 42(3): 135–147.
References 12. Gras L-L, Mitton D, Viot P, et al. Hyper-elastic proper-
1. Payne-James J, Jones R, Karch SB, et al. Simpson’s for- ties of the human sternocleidomastoideus muscle in ten-
ensic medicine. 13th ed. Boca Raton, FL: CRC Press, sion. J Mech Behav Biomed Mater 2012; 15: 131–140.
2011. 13. Amaral TF, Restivo MT, Guerra RS, et al. Accuracy of
2. Saukko P and Knight B. Knight’s forensic pathology. 3rd a digital skinfold system for measuring skinfold thickness
ed. Boca Raton, FL: CRC Press, 2004. and estimating body fat. Br J Nutr 2011; 105(03):
3. DiMaio D and DiMaio VJM. Forensic pathology: practi- 478–484.
cal aspects of criminal and forensic investigations. 2nd ed. 14. Chouzal MF, Restivo MT, Quintas MR, et al. Adips-
Boca Raton, FL: CRC Press, 2001. meter: a new skinfold calliper system. In: BIODEVICES
4. Prahlow JA. Forensic pathology for police, death investi- 2011 proceedings of the international conference on biome-
gators, attorneys, and forensic scientists. Totowa, NJ: dical electronics and devices, Rome, Italy, 26–29 January
Humana Press, 2010. 2011, pp.174–178. SciTePress, http://dblp.uni-trier.de/
5. Kobayashi M, Takatori T, Iwadate K, et al. Reconsidera- rec/bibtex/conf/biostec/ChouzalRQSAA11
tion of the sequence of rigor mortis through postmortem 15. Restivo MT, Amaral TF, Chouzal MF, et al. A digital
changes in adenosine nucleotides and lactic acid in differ- calliper for training and study purposes. Asia Pac J Clin
ent rat muscles. Forensic Sci Int 1996; 82(3): 243–253. Nutr 2012; 21(2): 182–190.
6. Van den Oever R. A review of the literature as to the 16. Spitzer V, Ackerman MJ, Scherzinger AL, et al. The visi-
present possibilities and limitations in estimating the time ble human male: a technical report. J Am Med Inform
of death. Med Sci Law 1976; 16(4): 269–276. Assoc 1996; 3(2): 118–130.
7. Madea B and Kernbach-Wighton G. Early and late post- 17. Tran H, Charleux F, Rachik M, et al. In vivo characteri-
mortem changes. In: Siegel JA and Saukko PJ (eds) zation of the mechanical properties of human skin
Encyclopedia of forensic sciences. Waltham, MA: Aca- derived from MRI and indentation techniques. Comput
demic Press, 2013, pp.217–228, http://www.sciencedirect.- Methods Biomech Biomed Engin 2007; 10(6): 401–407.
com/science/article/pii/B9780123821652001872 18. Holzapfel GA. Nonlinear solid mechanics: a continuum
8. Dix J and Graham M. Time of death, decomposition and approach for engineering. Chichester: John Wiley & Sons,
identification: an atlas (cause of death atlas series). Boca 2000.
Raton, FL: CRC Press, 1999, p.10. 19. Evans S and Holt C. Measuring the mechanical proper-
9. Fung YC. A first course in continuum mechanics. 3rd ed. ties of human skin in vivo using digital image correlation
Upper Saddle River, NJ: Prentice Hall, 1994. and finite element modelling. J Strain Anal Eng 2009;
44(5): 337–345.