Jean-Sébastien Raul

Finite-element models of the human head and their applications in forensic practice

Since the 1960s, predictive human head impact indices have been developed to help the investigation of causation of human head injury. Finite-element models (FEM) can provide interesting tools for the forensic scientists when various human head injury mechanisms need to be evaluated. Human head FEMs are mainly used for car crash evaluations and are not in common use in forensic science. Recent technological progress has resulted in creating more simple tools, which will certainly help to consider the use of FEM in routine forensic practice in the coming years. This paper reviews the main FEMs developed and focuses on the models which can be used as predictive tools. Their possible applications in forensic medicine are discussed.

Finite element modelling of human head injuries caused by a fall

Finite element models (FEMs) can be used as prediction tools for human head injuries caused by falls. The purpose of this paper is to demonstrate the relevance of using human head FEM to assess the possible mechanism for the origin of head injuries. The FEM of the human head used in this study was developed in the late 1990s at the University Louis Pasteur of Strasbourg (ULP) and has been validated for human head impacts for simulating human head injuries caused by car accidents. Its use in legal medicine appears to be very useful for comparing different injury mechanisms. We present the simulation obtained for two witnessed falls of the same individual, and compare our results to tolerance limits of the main human head injuries. We show that this tool can be used to discuss the possible mechanism of injury encountered for the observed lesions in a forensic case. It can also help to distinguish between possible and impossible human head injury mechanisms.

Finite element analysis of impact and shaking inflicted to a child.

This study compares a vigorous shaking and an inflicted impact, defined as the terminal portion of a vigorous shaking, using a finite element model of a 6-month-old child head. Whereas the calculated values in terms of shearing stress and brain pressure remain different and corroborate the previous studies based on angular and linear velocity and acceleration, the calculated relative brain and skull motions that can be considered at the origin of a subdural haematoma show similar results for the two simulated events. Finite element methods appear as an emerging tool in the study of the biomechanics of head injuries in children.

Biofidelic child head FE model to simulate real world trauma

Biomechanics of human head has been widely studied since several decades. At a mechanical level, the use of engineering allowed investigating injury mechanisms developing numerical models of adult head. For children, the problem is more difficult and evaluating child injury mechanisms using data obtained from scaling adult injury criteria does not account for differences in morphology and structure between adults and children. During growth, child head undergoes different modifications in morphology and structure. The present paper compares the anthropometry and numerical simulations of a child head model based on medical CT scans to a child head model developed by scaling an adult head model using the method proposed by Mertz [H.J. Mertz, A procedure for normalizing impact response data, SAE paper 840884, 1984]. These analysis point out significant differences showing that scaling down an adult head to obtain a child head does not appear relevant. Biofidelic and specific child geometry is needed to investigate child injury mechanisms.

Finite element modelling of paediatric head impact: Global validation against experimental data

Biomechanics of the human head has been widely studied for several decades. At a mechanical level, the use of engineering and finite element (FE) methods has allowed injury mechanisms to be investigated using biofidelic FE models. These models are generally validated using experimental data then used to simulate real-world head trauma in order to derive numerical tolerance limits, leading to efficient injury predicting tools. Due to ethical issues, experimental tests on the paediatric population remain prohibitive so direct validations of numerical models cannot be performed. However injury biomechanics on paediatric population is emerging with experimental tests on the paediatric cadavers or tests on biological tissue and the development of finite element models. The present paper proposes a new finite element model of a newborn head, simulating its main features, with material properties from the literature. Global validation of the model against experimental data in terms of skull deflection is performed and the model is used to simulate paediatric skull fracture coming from real-world head trauma.

Towards child versus adult brain mechanical properties.

The characterization of brain tissue mechanical properties is of crucial importance in the development of realistic numerical models of the human head. While the mechanical behavior of the adult brain has been extensively investigated in several studies, there is a considerable paucity of data concerning the influence of age on mechanical properties of the brain. Therefore, the implementation of child and infant head models often involves restrictive assumptions like properties scaling from adult or animal data. The present study presents a step towards the investigation of the effects of age on viscoelastic properties of human brain tissue from a first set of dynamic oscillatory shear experiments. Tests were also performed on three different locations of brain (corona radiata, thalamus and brainstem) in order to investigate regional differences. Despite the limited number of child brain samples a significant increase in both storage and loss moduli occurring between the age of 5 months and the age of 22 months was found, confirmed by statistical Students t-tests (p=0.104,0.038 and 0.054 for respectively corona radiata, thalamus and brain stem samples locations respectively). The adult brain appears to be 3-4 times stiffer than the young child one. Moreover, the brainstem was found to be approximately 2-3 times stiffer than both gray and white matter from corona radiata and thalamus. As a tentative conclusion, this study provides the first rheological data on the human brain at different ages and brain regions. This data could be implemented in numerical models of the human head, especially in models concerning pediatric population.

Influence of the benign enlargement of the subarachnoid space on the bridging veins strain during a shaking event: a finite element study

There is controversy regarding the influence of the benign enlargement of the subarachnoid space on intracranial injuries in the field of the shaken baby syndrome. In the literature, several terminologies exists to define this entity illustrating the lack of unicity on this theme, and often what is “benign” enlargement is mistaken with an old subdural bleeding or with abnormal enlargement due to brain pathology. This certainly led to mistaken conclusions. To investigate the influence of the benign enlargement of the subarachnoid space on child head injury and especially its influence on the bridging veins, we used a finite element model of a 6-month-old child head on which the size of the subarachnoid space was modified. Regarding the bridging veins strain, which is at the origin of the subdural bleeding when shaking an infant, our results show that the enlargement of the subarachnoid space has a damping effect which reduces the relative brain/skull displacement. Our numerical simulations suggest that the benign enlargement of the subarachnoid space may not be considered as a risk factor for subdural bleeding.

Tissue specificity and regulation of the N-terminal diversity of reticulon 3

Over the last few years, the widely distributed family of reticulons (RTNs) is receiving renewed interest because of the implication of RTN4/Nogo in neurite regeneration. Four genes were identified in mammals and are referred to as RTN1, 2, 3 and the neurite outgrowth inhibitor RTN4/Nogo. In the present paper, we describe the existence of five new isoforms of RTN3 that differ in their N-termini, and analysed their tissue distribution and expression in neurons. We redefined the structure of human and murine rtn3 genes, and identified two supplementary exons that may generate up to seven putative isoforms arising by alternative splicing or differential promoter usage. We confirmed the presence of five of these isoforms at the mRNA and protein levels, and showed their preferential expression in the central nervous system. We analysed rtn3 expression in the cerebellum further, and observed increased levels of several of the RTN3 isoforms during cerebellum development and during in vitro maturation of cerebellar granule cells. This pattern of expression paralleled that shown by RTN4/Nogo isoforms. Specifically, RTN3A1 expression was down-regulated upon cell death of cerebellar granule neurons triggered by potassium deprivation. Altogether, our results demonstrate that the rtn3 gene generates multiple isoforms varying in their N-termini, and that their expression is tightly regulated in neurons. These findings suggest that RTN3 isoforms may contribute, by as yet unknown mechanisms, to neuronal survival and plasticity.

Dehydroepiandrosterone in nails of infants: a potential biomarker of intrauterine responses to maternal stress.

Easily accessible biomarkers for fetal stress biology are lacking. We here explore whether quantification of major fetal steroids, dehydroepiandrosterone (DHEA) or DHEA sulfate (DHEAS), with liquid chromatography/tandem mass spectrometry in infant nails is a tool to assess fetal stress biology in response to maternal stressful life events during pregnancy. Sufficient nail (≥ 1 mg) was available from 80 infants (93% of those providing samples). The concentration of DHEA, but not DHEAS, was increased in infants of mothers with stressful life events during pregnancy (DHEA: F₁,₄₁=6.105, P=0.018; DHEAS: F₁,₇₇=0.767, P=0.384). DHEA concentrations were not related to maternal stress before pregnancy (F₁,₄₁=0.010, P=0.922). Infant nail DHEA may be a fetal biological correlate of intrauterine exposure to maternal stress. The method promises the first non-invasive retrospective biomarker for intrauterine stress biology, opening new ways for research and clinical applications in fetal medicine, endocrinology, obstetrics, gynecology, and for understanding the developmental origins of health and disease.

Limitation of scaling methods in child head finite element modelling

During growth, a childs head undergoes different modifications in morphology and structure. This paper presents an anthropometric study in terms of dimension compared to the scaling method developed by Mertz which consists of reducing the adult head model with a scaling coefficient to obtain a child head. A detailed sizes and shape analysis of brain contour in sagittal and frontal plans is then proposed, for child head versus a scaled adult head. The superimposition of those contours allowed pointing to main differences. Numerical simulations performed with the detailed three year old child head model developed in this present study, and a scaled adult head finite element model, showed that reducing an adult finite element model to obtain a child head by scaling method does not seem to be realistic. Then, the creation of specific finite element models of child head seems necessary to understand paediatric injuries.