The evolution of methods for estimating the time of death from body temperature

A pilot study investigating early postmortem interval of rats based on ambient temperature and postmortem interval-related metabolites in blood

Estimation of the postmortem interval (PMI), especially the early PMI, plays a key role in forensic practice. Although several studies based on metabolomics approaches have presented significant findings for PMI estimation, most did not examine the effects of ambient temperature. In this study, gas chromatography-mass spectrometry (GC‒MS)‒based metabolomics was adopted to explore the changes in metabolites in the cardiac blood of suffocated rats at various ambient temperatures (5 °C, 15 °C, 25 °C, and 35 °C) from 0 to 24 h after death. Isoleucine, alanine, proline, valine, glycerol, glycerol phosphate, xanthine, and hypoxanthine were found to contribute to PMI in all temperature groups. Hypoxanthine and isoleucine were chosen to establish estimation models (equations) with an interpolation function using PMI as the dependent variable (f(x, y)), relative intensity as the independent variable x, and temperature as the independent variable y. Thereafter, these two models were validated with predictive samples and shown to have potential predictive ability. The findings indicate that isoleucine, alanine, proline, valine, glycerol, glycerol phosphate, xanthine, and hypoxanthine may be significant for PMI estimation at various ambient temperatures. Furthermore, a method to determine PMI based on ambient temperature and PMI-related metabolites was explored, which may provide a basis for future studies and practical applications.

Beyond Henssge’s Formula: Using Regression Trees and a Support Vector Machine for Time of Death Estimation in Forensic Medicine

Henssge’s nomogram is a commonly used method to estimate the time of death. However, uncertainties arising from the graphical solution of the original mathematical formula affect the accuracy of the resulting time interval. Using existing machine learning techniques/tools such as support vector machines (SVMs) and decision trees, we present a more accurate and adaptive method for estimating the time of death compared to Henssge’s nomogram. Using the Python programming language, we built a synthetic data-driven model in which the majority of the selected tools can estimate the time of death with low error rates even despite having only 3000 training cases. An SVM with a radial basis function (RBF) kernel and AdaBoost+SVR provided the best results in estimating the time of death with the lowest error with an estimated time of death accuracy of approximately ±20 min or ±9.6 min, respectively, depending on the SVM parameters. The error in the predicted time (tp[h]) was tp±0.7 h with a 94.45% confidence interval. Because training requires only a small quantity of data, our model can be easily customized to specific populations with varied anthropometric parameters or living in different climatic zones. The errors produced by the proposed method are a magnitude smaller than any previous result.

Case closed - Wrappings and encasement delays and reduces fly presence on body parts

Homicide, particularly where a body has been concealed, is uniquely challenging for investigators to estimate the time of occurrence due to the methods employed by perpetrators to hide the body or its constituent parts from detection. The regularity of necrophagous insect lifecycles to determine minimum post-mortem interval (minPMI) is widely employed but remains an unreliable technique if used without a clear understanding of the factors that affect insect access and oviposition behaviour to concealed remains. The purpose of this study was to investigate the effect of wrapping body parts on fly colonisation and implications for minPMI calculations. Field studies were carried out using four treatments of pork (as surrogate body parts), in five replicates, one unwrapped, the other three wrapped in either a black plastic sack, a small-zipped wash bag (to simulate a suitcase), or a plastic sack further placed in a wash bag. Over a 48-h period all the methods of wrapping significantly disrupted the host-finding process of blowflies to dismembered carcasses, with a delay of initial contact and oviposition of 30+h (dependant on wrapping) and even more in wet conditions (48+ h). Egg numbers were also reduced by as much as 99.1% on wrapped samples compared to unwrapped. These new findings highlight the importance of applying adjustments to minPMI calculations when encountering wrapped remains. Advances in the accuracy of minPMI calculations will prevent the waste of valuable police time and resources and better focus the search for witnesses and suspects in homicide investigations.

Individualised and non-contact post-mortem interval determination of human bodies using visible and thermal 3D imaging

Determining the time since death, i.e., post-mortem interval (PMI), often plays a key role in forensic investigations. The current standard PMI-estimation method empirically correlates rectal temperatures and PMIs, frequently necessitating subjective correction factors. To overcome this, we previously developed a thermodynamic finite-difference (TFD) algorithm, providing a rigorous method to simulate post-mortem temperatures of bodies assuming a straight posture. However, in forensic practice, bodies are often found in non-straight postures, potentially limiting applicability of this algorithm in these cases. Here, we develop an individualised approach, enabling PMI reconstruction for bodies in arbitrary postures, by combining photogrammetry and TFD modelling. Utilising thermal photogrammetry, this approach also represents the first non-contact method for PMI reconstruction. The performed lab and crime scene validations reveal PMI reconstruction accuracies of 0.26 h ± 1.38 h for true PMIs between 2 h and 35 h and total procedural durations of ~15 min. Together, these findings broaden the potential applicability of TFD-based PMI reconstruction.

Establishing the time since death (TSD) is vital in many forensic investigations. By combining thermometry, photogrammetry and numerical thermodynamic modelling, the TSD can be determined non-invasively for bodies of arbitrary shape and posture with an unprecedented accuracy of 0.26 h ± 1.38 h.

Infrared thermography as a modality for tracking cutaneous temperature change and post-mortem interval in the critical care setting

PURPOSE

To evaluate the potential use of cutaneous facial temperature change as measured by an infrared camera as a marker of postmortem interval (PMI) in the minutes immediately following death.

METHODS

This was a prospective, observational pilot study using a convenience sample of all deaths which occurred in a room in an Intensive Care Unit equipped with a ceiling mounted thermal camera. Cutaneous temperature measurements were taken from 60 min antemortem to as long as possible postmortem.

RESULTS

A total of 134 separate measurements was taken from 5 patients, with 65 occurring antemortem, and 69 occurring post-mortem. The longest recorded post-mortem time was 130 min. A Kruskal-Wallis ANOVA testing the hypothesis that there was a difference in facial temperature at each of the different timepoints showed significance (p = 0.029). Post-Hoc comparisons were then performed to compare median temperature values at each timeframe to the baseline value. Compared to baseline, there was a significant difference in facial temperature at 30, 60, and 90 min (p = 0.007, p = 0.01, p = 0.016) (Table 2).

CONCLUSION

There is a statistically significant cutaneous facial temperature change in patients immediately following death as measured by a thermal camera. There is potential for infrared thermography to identify changes immediately before and after death in environments where traditional temperature measurement cannot be accomplished. More work needs to be done to confirm whether a precise postmortem interval (PMI) could be derived from these values.

Reconstructing the time since death using noninvasive thermometry and numerical analysis

The early postmortem interval (PMI), i.e., the time shortly after death, can aid in the temporal reconstruction of a suspected crime and therefore provides crucial information in forensic investigations. Currently, this information is often derived from an empirical model (Henssge's nomogram) describing posthumous body cooling under standard conditions. However, nonstandard conditions necessitate the use of subjective correction factors or preclude the use of Henssge's nomogram altogether. To address this, we developed a powerful method for early PMI reconstruction using skin thermometry in conjunction with a comprehensive thermodynamic finite-difference model, which we validated using deceased human bodies. PMIs reconstructed using this approach, on average, deviated no more than ±38 minutes from their corresponding true PMIs (which ranged from 5 to 50 hours), significantly improving on the ±3 to ±7 hours uncertainty of the gold standard. Together, these aspects render this approach a widely applicable, i.e., forensically relevant, method for thermometric early PMI reconstruction.

An experimental study on investigating the postmortem interval in dichlorvos poisoned rats by GC/MS-based metabolomics

The estimation of the postmortem interval (PMI) is always a key issue in forensic science. Although many attempts based on metabolomics approaches have been proven to be feasible and accurate for PMI estimation, there have been no reports regarding the determination of the PMI in acute dichlorvos (DDVP) poisoning. In this study, all rats were killed by acute DDVP poisoning at a dose three fold the oral LD₅₀ (240 mg/kg). Gas chromatography-mass spectrometry (GC/MS) was applied to investigate the metabolic profiling of blood samples at various times after death up to 72 h. A total of 39 metabolites were found to be associated with PMI, and the combinations of various numbers of metabolites were used to establish support vector regression (SVR) models to investigate the PMI. The SVR model constructed by 23 metabolites had a minimum mean squared error (MSE) of 5.49 h for the training set. Then, the SVR model was validated by prediction set with an MSE of 10.33 h, suggesting good predictive ability of the model for investigating the PMI. The findings demonstrated the great potential of GC/MS-based metabolomics combined with the SVR model in determining the PMI of DDVP poisoned rats and provided an experimental basis for the application of this approach in investigating the PMI of other toxicants.

Technical note: early post-mortem changes of human bone in taphonomy with μCT

Post-mortem interval (PMI) estimation is an important issue in forensic medicine, particularly for criminal purposes and legal limitation periods. The goal of the present study is to examine the evolution of the trabecular cranial vault bone after 4 weeks of conservation in a controlled environment with micro-tomography (μCT) analyses.Four bone samples were extracted from a fresh human cranial vault (a donation to science according to the French law) and conserved in an air-controlled environment. The samples were weighed and μCT scanned at a 10-μm resolution every week after death for a month. The μCT features were identical for every sample. Each set of data from the μCTs was reconstructed, registered, and analyzed in terms of the total volume, bone volume, bone surface, number of trabeculae, trabeculae thickness, and mean distance of the trabeculae. The samples were conserved in a glass box in 20 °C air with 60% humidity in a laboratory hood between each μCT acquisition. Descriptive statistics were determined. Each sample was observed and compared to itself over time.After 1 month of conservation, the mean bone volume (-1.9%), bone surface (-5.1%), and trabecular number (-12.35%) decreased, whereas the mean trabecular separation (+5.55%) and trabecular thickness (+12.7%) increased. Many variations (i.e., increases and decreases) were observed between the extraction of the sample and the end of the 4 weeks of conservation. The present observations may be explained by bone diagenesis. Previous observations have indicated that protein and lipid losses occur with bone weight and volume losses. These diagenesis effects may explain the trabecular modifications observed in the present work. We observed many bone variations with the μCT scans between the beginning and the end of the conservation that had no explanations. Additional studies, particularly studies involving statistics, need to be performed to confirm our observations and explain these results more clearly.

Short-term effects of hydrated lime and quicklime on the decay of human remains using pig cadavers as human body analogues: Laboratory experiments

Contradictions and misconceptions regarding the effect of lime on the decay of human remains have demonstrated the need for more research into the effect of different types of lime on cadaver decomposition. This study follows previous research by the authors who have investigated the effect of lime on the decomposition of human remains in burial environments. A further three pig carcasses (Sus scrofa), used as human body analogues, were observed and monitored for 78 days without lime, with hydrated lime (Ca(OH)2) and with quicklime (CaO) in the taphonomy laboratory at the University of Bradford. The results showed that in the early stages of decay, the unlimed and hydrated lime cadavers follow a similar pattern of changes. In contrast, the application of quicklime instigated an initial acceleration of decay. Microbial investigation demonstrated that the presence of lime does not eliminate all aerobic bacteria. The experiment also suggested that lime functions as a sink, buffering the carbon dioxide evolution. This study complements the field observations. It has implications for the investigation of time since death of limed remains. Knowledge of the effects of lime on decomposition processes is of interest to forensic pathologists, archaeologists, humanitarian organisations and those concerned with disposal of animal carcasses or human remains in mass disasters.

Some challenges in forensic veterinary pathology: a review

Forensic veterinary pathology is a diverse discipline that is in an early phase of its development. Common challenges include estimation of the age of skin wounds and bruises, the diagnosis of drowning and estimation of the time since death. However, many details of the pathological findings related to these various aspects await validation. The 'multispecies' nature of veterinary pathology, combined with the preponderance of published observations originating from animal experimentation, rather than casework, poses two challenges. Firstly, extrapolation of results between species may jeopardize the reliability (and credibility) of the forensic opinion. Secondly, experimental studies may not truly reflect the spectrum of changes seen in actual cases (e.g. extent of injuries, infection, age and health of victim). With regard to drowning, diagnosis based on post-mortem findings remains problematical. Methods for estimation of the time since death (also known as the post-mortem interval) continue to be a major focus of study, with fresh avenues such as post-mortem diagnostic imaging offering interesting possibilities.

Estimation of early postmortem intervals by a multiple regression analysis using rectal temperature and non-temperature based postmortem changes

Five general methods based on rectal temperature and a multiple regression analysis using rectal temperature and non-temperature based postmortem changes were applied to 212 postmortem cases of within 24h postmortem (PM) intervals. Non-temperature based postmortem changes of rigidity, hypostasis and corneal turbidity were numerically categorized and used with rectal temperatures as four statistical variables in the multiple regression analysis. The correlation coefficient values between true and calculated postmortem intervals were 0.78-0.82 in the five general methods based on rectal temperature. The multiple regression analysis produced a multiple correlation coefficient value of 0.89 and according to the error ranges of the PM intervals, 72% of the cases were estimated within the error of +/-1.0 h and 92% within +/-5.0 h. Although assessments of non-temperature based PM changes are mostly subjective and have a wide variation, the present study demonstrated a usefulness of non-temperature based PM changes in the estimation of PM intervals.

Comparison of morphological changes in white blood cells after death and in vitro storage of blood for the estimation of postmortem interval

Estimation of the time of death is one of the most important problems for forensic medicine and law. Physical and chemical postmortem changes are evaluated together while estimating the time of death. In this study, in vitro storage and postmortem changes of white blood cells were aimed to be compared within the given postmortem interval, and a follow-up study was carried out. Blood smears which were obtained from 10 non-refrigerated cadavers (experimental group) and from 40 hospital patients (control group) have been evaluated to observe and compare changes during the in vitro storage and postmortem degenerative morphological changes that white blood cells undergo throughout the given postmortem intervals. The samples were examined by using a light microscope, and blood cells were differentiated by staining blood films with May-Grunwald stain, followed by Giemsa stain. Identifiable degenerated eosinophils and monocytes were first examined at 6h of death and the in vitro storage, and they were unidentifiable beyond 72 h of storage. Identifiable degeneration of neutrophils were first examined at 6h of death and storage while unidentifiable beyond 96 h of storage. Identifiable degeneration of lymphocytes were first examined at 24h of death, and they were still identifiable beyond 120 h. Cellular changes of leukocytes can be useful in the 6-120 h for estimating the time of in vitro storage, and the findings match during the first 21 h for both experimental and control groups. Finally, this follow-up study and the comparison will also be carried out for a longer postmortem interval, and other specific hypothesis that relate cellular changes in tissues other than blood with time since death are various points that needs to be studied.

Outer ear temperature and time of death

From a research sample of 138 corpses, divided into four subgroups of ambient storage temperature (0-5 degrees C, 6-10 degrees C, 11-15 degrees C and 16-23 degrees C) four linear regression formulae of actual versus estimated post-mortem interval were obtained ('interval' formulae) using a single outer ear temperature measurement on both sides. This method showed the best correlation coefficient among five other methods previously proposed for time of death determination (rectal temperature, vitreous K+, CSF K+, blood log NA+/K+ and log Cl-), however its results were less accurate than those obtained with a multivariate equation combining several of the above mentioned methods. Eventually an equation expressing time of death (TOD) as a function of outer ear temperature (OE T degrees) and ambient temperature was also established from the whole research sample ('global' formulae). On a different sample of 141 corpses the regression formulae ('interval' and 'global') for the outer ear temperature were compared to three methods based on a single rectal temperature measurement ('rule of thumb' 1 and 2, Henssge nomogram) and therefore useful at the scene; the results of all methods were compared within the four subgroups of ambient temperature as well as in three subgroups of different post-mortem interval lengths (< 7 h, < 10 h, < 15 h). In all cases the outer ear temperature formulae provided better results than the rectal temperature methods (especially Henssge nomogram and rule of thumb 1). Moreover they did not show any post-mortem plateau which was present in almost 30% of cases when rectal temperature was measured in corpses kept at ambient temperature above 15 degrees C. Our results show that outer ear temperature measurement is the method which provides the best simplicity/quality ratio and should therefore be proposed for use at the scene when conditions are similar to those of our experiment (within buildings). A software equipped thermometer is required in order to use in each case the appropriate formula and confidence interval.

The application of eight reported temperature-based algorithms to calculate the postmortem interval

There exist numerous temperature-based techniques to calculate the early postmortem interval of human corpses. This paper presents eight commonly used time of death algorithms and describes how they were applied to eight corpses.

A new biochemical method for estimation of postmortem time

Hypoxanthine (Hx) is formed by hypoxic degradation of adenosine monophosphate (AMP) and might be elevated due to antemortem hypoxia. However, it also increases after cessation of the life processes. Until now measurements of potassium in corpus vitreous humor have been used by forensic pathologists to determine postmortem time. In this study the influence of postmortem time and temperature on vitreous humor Hx and potassium levels were compared. Repeated sampling of vitreous humor was performed in 87 subjects with known time of death and diagnosis. The bodies were kept at either 5 degrees C, 10 degrees C, 15 degrees C or 23 degrees C. Hx was measured by means of HPLC and potassium by flame photometry. In 19 subjects from whom samples were obtained within 1.5 h after death, the normal level of Hx could be estimated to be 7.6 mumol/l and that of potassium to be 5.8 mmol/l. The spread of the potassium levels measured shortly after death was much greater than for the corresponding Hx levels. In the four temperature groups the Hx level increased 4.2, 5.1, 6.2 and 8.8 mumol/l per h, respectively, whereas the corresponding figures for potassium were 0.17, 0.20, 0.25 and 0.30 mmol/l per h. The vitreous humor concentration of both Hx and potassium increases fairly linearly after death. The slopes are steeper with increasing temperature. Since the scatter of the levels is greater for potassium than for Hx, the latter parameter seems to be better suited for the determination of time of death in cases without antemortem hypoxia, especially during the first 24 h.

Magnesium, calcium and zinc fluctuations on skin induced injuries in correlation with time of induction

In the present study, fluctuations of trace elements Mg, Ca and Zn concentrations with time on skin-induced injuries was investigated. To accomplish this, 144 animals (pig) aged between 5-6 months scheduled for food provision process (slaughter) were used. At the gluteus area, injuries were induced prior to slaughter at intervals of 30 s, 30 min, 1, 2, 4, 8 h. Local anaesthesia and cold therapy for prompt relief of pain (ethylchloride, C2H5Cl) was applied. Postmortem tissue excision in and around the injured site was promptly performed. The tissue obtained was segmented into three zones of equal distances (2 cm) in between and weighed 3 g wet weight. Tissue specimens were analysed by atomic absorption spectroscopy for the three elements. The results were correlated with time of injury. Suggestive alterations in trace elements mean concentrations with time were confirmed. The ratios of the mean in twos (Ca/Zn, Ca/Mg, Mg/Zn) versus time were graphed. Effectively, the curves achieved by analysing postmortem tissues, serve to estimate the time of an injury induced in vivo.

Estimation of the post mortem period by multiple-site temperature measurements and the use of a new algorithm

The estimation of the post mortem period, in cases where death occurred under suspicious circumstances, is usually attempted using temperature measurements taken at a single body site. Early investigations of the validity of such an approach use the abdominal skin surface, the axilla and the rectum as measurement sites (B.H. Knight, Forensic Sci. Int., 36 (1987) 47-55). However, it has recently become more common to use the rectum alone, though the ear and the nasal passages have also been utilized. Whatever site is employed, the estimates are frequently found to be inaccurate. There are several fundamental reasons for these inaccuracies, the most prominent being the unknown variation in the ambient temperature between the time of death and the commencement of measurements, and the unknown body temperature at the time of death. This paper proposes a method of overcoming the above difficulties by taking a series of measurements concurrently at a number of body sites, a technique used by several previous workers (B.H. Knight, Forensic Sci. Int., 36 (1987) 47-55). Initial investigations have shown that an improved estimation of post mortem period is obtainable by the application of a suitable decision-making algorithm.