Refine
Document Type
- Article (6)
- ZIB-Report (2)
- Research data (1)
Language
- English (9)
Is part of the Bibliography
- no (9)
Keywords
Temperature based death time estimation is based either on simple phenomenological models of corpse cooling or on detailed physical heat transfer models. The latter are much more complex, but allow a higher accuracy of death time estimation as in principle all relevant cooling mechanisms can be taken into account. Here, a complete work flow for finite element based cooling simulation models is presented.
The following steps are demonstrated on CT-phantoms:
• CT-scan
• Segmentation of the CT images for thermodynamically relevant features of individual
geometries
• Conversion of the segmentation result into a Finite Element (FE) simulation model
• Computation of the model cooling curve
• Calculation of the cooling time
For the first time in FE-based cooling time estimation the steps from the CT image over segmentation to FE model generation are semi-automatically performed. The cooling time calculation results are compared to cooling measurements performed on the phantoms under controlled conditions. In this context, the method is validated using different CTphantoms. Some of the CT phantoms thermodynamic material parameters had to be experimentally determined via independent experiments. Moreover the impact of geometry and material parameter uncertainties on the estimated cooling time is investigated by a sensitivity analysis.
Estimation of time of death based on a single measurement of body
core temperature is a standard procedure in forensic medicine.
Mechanistic models using simulation of heat transport promise
higher accuracy than established phenomenological models in
particular in nonstandard situations, but involve many not exactly
known physical parameters. Identifying both time of death and
physical parameters from multiple temperature measurements is
one possibility to reduce the uncertainty significantly.
In this paper, we consider the inverse problem in a Bayesian setting
and perform both local and sampling-based uncertainty
quantification, where proper orthogonal decomposition is used as
model reduction for fast solution of the forward model. Based on
the local uncertainty quantification, optimal design of experiments
is performed in order to minimize the uncertainty in the time of
death estimate for a given number of measurements. For reasons
of practicability, temperature acquisition points are selected from
a set of candidates in different spatial and temporal locations.
Applied to a real corpse model, a significant accuracy improvement
is obtained already with a small number of measurements.
Estimation of time of death based on a single measurement of body
core temperature is a standard procedure in forensic medicine.
Mechanistic models using simulation of heat transport promise
higher accuracy than established phenomenological models in
particular in nonstandard situations, but involve many not exactly
known physical parameters. Identifying both time of death and
physical parameters from multiple temperature measurements is
one possibility to reduce the uncertainty significantly.
In this paper, we consider the inverse problem in a Bayesian setting
and perform both local and sampling-based uncertainty
quantification, where proper orthogonal decomposition is used as
model reduction for fast solution of the forward model. Based on
the local uncertainty quantification, optimal design of experiments
is performed in order to minimize the uncertainty in the time of
death estimate for a given number of measurements. For reasons
of practicability, temperature acquisition points are selected from
a set of candidates in different spatial and temporal locations.
Applied to a real corpse model, a significant accuracy improvement
is obtained already with a small number of measurements.
Temperature-based estimation of time of death (ToD) can be per-
formed either with the help of simple phenomenological models of corpse
cooling or with detailed mechanistic (thermodynamic) heat transfer mod-
els. The latter are much more complex, but allow a higher accuracy of
ToD estimation as in principle all relevant cooling mechanisms can be
taken into account.
The potentially higher accuracy depends on the accuracy of tissue and
environmental parameters as well as on the geometric resolution. We in-
vestigate the impact of parameter variations and geometry representation
on the estimated ToD based on a highly detailed 3D corpse model, that
has been segmented and geometrically reconstructed from a computed to-
mography (CT) data set, differentiating various organs and tissue types.
Temperature-based estimation of time of death (ToD) can be per-
formed either with the help of simple phenomenological models of corpse
cooling or with detailed mechanistic (thermodynamic) heat transfer mod-
els. The latter are much more complex, but allow a higher accuracy of
ToD estimation as in principle all relevant cooling mechanisms can be
taken into account.
The potentially higher accuracy depends on the accuracy of tissue and
environmental parameters as well as on the geometric resolution. We in-
vestigate the impact of parameter variations and geometry representation
on the estimated ToD based on a highly detailed 3D corpse model, that
has been segmented and geometrically reconstructed from a computed to-
mography (CT) data set, differentiating various organs and tissue types.
From that we identify the most crucial parameters to measure or estimate,
and obtain a local uncertainty quantifcation for the ToD.
Code and data for the reconstruction of ambient temperature drop in time of death estimation
We provide Octave code and temperature measurement data for
- empirircally estimating thermal sensor likelihood
- estimating time and amplitude of a single sudden ambient temperature drop from
temperature measurement data in two thermally different compartments.
In medicolegal practice, time since death is estimated to assess alibi for homicide cases. Ambient temperature TA has a strong impact on cooling and therefore on temperature based time since death estimation (TTDE). At many crimescenes the ambient temperature TA1 is lowered instantaneously from a start value TA0 to a value TA1 at a certain time t0 during investigations due to human intervention such as window or door opening or body transport. Usually TA0 and t0 are unknown to the investigators. In this paper we focus on reconstruction of the unknown parameters TA0 and t0.
Our approach is inspired by TTDE literature remarks of detecting said changes by measuring temperatures in closed compartments as e.g. cupboards or neighboring rooms of the crime scene, where TA0 could have been ‘preserved’ after t0. We aim to estimate t0 and TA0 from temperature measurements TZ(t) in closed compartments Z at times t > t0.
We got results even under the most trivial assumption of Newtonian cooling for boxes filled with air, with heaps of clothes or even with books in two different experimental scenarios. Two different parameter estimators, (t0^, TA0^) using a single quadruple temperature measurement in two boxes and (t0*, TA0*) on the basis of weighted averaging the results of a series of N quadruple measurements during cooling of the two boxes respectively, were tested.
Our results were partially appropriate for TTDE input. For example a sudden decline at time t0 from TA0 = 22.5°C to TA1 = 14°C of the ambient temperature in a climate chamber could be reconstructed at t = t0 + 95min with relative deviations ρt0^ = 27% and ρTA0^ = 19% of the estimators relative to t - t0 and TA0 – TA1 respectively, only based on N = 1 quadruple measurement with a span of Δt = 50min. In case of N = 200 quadruple measurements starting at t = t0 + 95min and ending at t = t0 + 295min we found for weighted mean estimators distinctively reduced relative deviations ρt0^ = 5% and ρTA0^ = 11% with the same quadruple span Δt = 50min.
Further research is necessary to guarantee applicability in routine case work. We will investigate more elaborate cooling models, estimation algorithms and evaluation localization.
Rectal temperature measurement (RTM) from crime scenes is an important parameter for temperature-based time of death estimation (TDE). Various influential variables exist in TDE methods like the uncertainty in thermal and environmental parameters. Although RTM depends in particular on the location of measurement position, this relationship has never been investigated separately. The presented study fills this gap using Finite Element (FE) simulations of body cooling. A manually meshed coarse human FE model and an FE geometry model developed from the CT scan of a male corpse are used for TDE sensitivity analysis. The coarse model is considered with and without a support structure of moist soil. As there is no clear definition of ideal rectal temperature measurement location for TDE, possible variations in RTM location (RTML) are considered based on anatomy and forensic practice. The maximum variation of TDE caused by RTML changes is investigated via FE simulation. Moreover, the influence of ambient temperature, of FE model change and of the models positioning on a wet soil underground are also discussed. As a general outcome, we notice that maximum TDE deviations of up to ca. 2-3 h due to RTML deviations have to be expected. The direction of maximum influence of RTML change on TDE generally was on the line caudal to cranial.
Temperature-based time of death estimation (TTDE) using simulation methods such as the finite element (FE) method promises higher accuracy and broader applicability in nonstandard cooling scenarios than established phenomenological methods. Their accuracy depends crucially on the simulation model to capture the actual situation. The model fidelity in turn hinges on the representation of the corpse’s anatomy in form of computational meshes as well as on the thermodynamic parameters.
While inaccuracies in anatomy representation due to coarse mesh resolution are known to have a minor impact on the estimated time of death, the sensitivity with respect to larger differences in the anatomy has so far not been studied. We assess this sensitivity by comparing four independently generated and vastly different anatomical models in terms of the estimated time of death in an identical cooling scenario. In order to isolate the impact of shape variation, the models are scaled to a reference size, and the possible impact of measurement location variation is excluded explicitly, which gives a lower bound on the impact of anatomy on the estimated time of death.