基于3D激光扫描、多刚体重建和遗传算法优化的车-人碰撞事故再现模拟与损伤分析

目的利用3D激光扫描技术,结合多刚体动力学和遗传算法优化对一起汽车与自行车及其骑车人碰撞的交通事故进行再现模拟与损伤分析,为法医学死因鉴定提供生物力学依据。方法利用3D激光扫描技术对肇事汽车进行测量,根据测量数据建立人-自行车-汽车多刚体模型,并设置优化变量的取值范围,利用多目标遗传算法,带精英策略的快速非支配排序遗传算法求最优近似解,并与事故现场附近监控视频记录进行比较。结果事故车辆激光扫描重建结果良好,通过遗传算法优化收敛所获得的最优近似解中的假人、自行车、汽车的动力学行为与监控视频中所记录的三者动力学行为相符合,假人的损伤参数也与事故中骑车人的损伤情况及部位相一致。结论利用3D激光扫描技术,结合多刚体动力学和遗传算法优化可以对人-自行车-汽车碰撞前的运动状态、碰撞致伤过程进行重建,并对受害者的损伤进行力学分析,在交通事故受害人致伤方式鉴定及死因分析中具有应用价值。     

Reconstruction of a real car-pedestrian-accident. Part I: Generating an individual digital topographic image of the victim

The aim of forensic biomechanics is the reconstruction of traumatic events based on the pathological findings in the victim's morphology, the accident traces and the car damages. The use of forensic documentation tools (e.g. Streifenlichttopometrie) enables 3-dimensional and proportional accurate documentation of the victim's body, of its injuries and of the car damages with submillimeter precision. The generated topographic image serves as input for a multi-body system model of the victim. It allows further to determine exactly the contact points between car and victim for a computer simulated dynamical reconstruction of the impact situation. In the case of an accident involving a car and a pedestrian the generation and application of computer aided 3-dimensional reconstruction models are shown.     

Application of 3D documentation and geometric reconstruction methods in traffic accident analysis: with high resolution surface scanning, radiological MSCT/MRI scanning and real data based animation

The examination of traffic accidents is daily routine in forensic medicine. An important question in the analysis of the victims of traffic accidents, for example in collisions between motor vehicles and pedestrians or cyclists, is the situation of the impact. Apart from forensic medical examinations (external examination and autopsy), three-dimensional technologies and methods are gaining importance in forensic investigations. Besides the post-mortem multi-slice computed tomography (MSCT) and magnetic resonance imaging (MRI) for the documentation and analysis of internal findings, highly precise 3D surface scanning is employed for the documentation of the external body findings and of injury-inflicting instruments. The correlation of injuries of the body to the injury-inflicting object and the accident mechanism are of great importance. The applied methods include documentation of the external and internal body and the involved vehicles and inflicting tools as well as the analysis of the acquired data. The body surface and the accident vehicles with their damages were digitized by 3D surface scanning. For the internal findings of the body, post-mortem MSCT and MRI were used. The analysis included the processing of the obtained data to 3D models, determination of the driving direction of the vehicle, correlation of injuries to the vehicle damages, geometric determination of the impact situation and evaluation of further findings of the accident. In the following article, the benefits of the 3D documentation and computer-assisted, drawn-to-scale 3D comparisons of the relevant injuries with the damages to the vehicle in the analysis of the course of accidents, especially with regard to the impact situation, are shown on two examined cases.     

The Relationship Between Deprivation and Forensic Material Recovered from Stolen Vehicles: Is it Affected by Vehicle Condition and Tidiness?

Previous research has shown that as crime scene location deprivation increases (lower socioeconomic status), the recovery of forensic material, principally DNA and fingerprints, also increases. However, this increase does not result in more crimes being solved by forensic means. In this study, we analyze stolen vehicle data and find a statistically significant positive association between deprivation and the amount of forensic material that matched either the victim or an associate of the victim on a criminal database. The nature of this association was investigated further by inspecting recovered stolen vehicles to establish whether the condition of a stolen vehicle and the tidiness of its interior influenced the recovery of forensic material that was attributed to the victim or an associate. Contradictory results suggest that other factors may contribute to understanding the association between the recovery of victim- or associate-attributable forensic material and crime scene location deprivation.     

Severity of vehicle bumper location in vehicle-to-pedestrian impact accidents

Pedestrian protection is one of the key topics for safety measures in traffic accidents all over the world. To analyze the relation between the collision site of the vehicle bumper and the severity of the lower extremity injuries, we performed biomechanical experiments. We compared the applied external force and the risks of subsequent injuries between the impact of the center and side positions of the front bumper. These comparisons were performed by practical impact tests with eight typical different types of cars which were typical of the current vehicle fleets. The tests were made using the TRL legform impactor which was a mechanical substitute of a pedestrian lower extremity. The TRL impactor is used all over the world for assessing the safety of car bumpers. It was found that the risks of lower extremity injuries in the impacts at the side positions, in front of the vehicle's side member, were significantly higher than those at the center. In the tests, we found that foam materials around the rigid front cross member had a significant effect on reducing the lower extremity injury risks and especially tibia fracture risk against vehicle bumper center collisions, but had little effect at the sides of the bumper over the vehicle's side members where the foam was thinner. We also found that the front shape of the vehicle affected the risk of ligaments injuries. According to these results, the information of impact locations of cars in vehicle-to-pedestrian traffic accidents is valuable for clinicians to diagnose patients with lower extremity injuries in traffic accidents and for forensic pathologists to analyze the accident reconstruction. Furthermore, the results suggest that testing of the bumper area in front of the main longitudinal beams should be included in the car safety legislation to require pedestrian safety.     

90例行人撞击伤案例分析

目的 探讨行人交通事故和致伤车辆类型的关系,总结其规律.方法 回顾90例行人交通事故案例,分析致伤车辆、损伤部位、损伤特点、严重程度之间的关系. 结果 不同类型致伤车辆所致的撞击伤好发部位取决于受撞击时人体的姿势和车头突出部位的高度,其损伤特点与作用力大小、方向和接触面光滑程度相关. 结论 通过对行人撞击伤的损伤部位、损伤特点、严重程度的分析,可以推断致伤车辆类型.     

道路交通事故死者MSCT扫描与尸体解剖结果的比较

目的使用多层螺旋CT(MSCT)对道路交通事故中死者的损伤进行观察分析,探索MSCT在道路交通事故鉴定中的应用价值。方法对10例道路交通事故死亡者进行全身MSCT扫描,随后进行系统尸体解剖,并将两者的结果进行对比,分析两者在获取交通事故损伤信息方面的优点和不足。结果 MSCT可以清晰显示身体各部位骨折三维形态特征,并可检查到全身各部位的积气,相对于尸体解剖优势明显;但对于器官和软组织损伤的检测,MSCT的分辨率有限。结论对于道路交通事故死者,确定死亡方式和死亡原因的最好方法是联合进行MSCT检查和尸体解剖。     

Traffic accident vs homicide: The contribution of DNA analysis to clarify this mystery. A case report

A marriage procession was going through the road when the vehicle met with a fatal accident and the wife of the driver died. The autopsy revealed lesions according with fatal traffic accident. But, a second autopsy revealed that there were injuries, but it was not reported in the first autopsy protocol. We analyze several autosomal STRs to typify some evidences collected inside the vehicle of traffic accident which were stained by the blood of the woman mortal victim. The results of the analysis of DNA suggested that the victim bled inside the vehicle and died and then, she was placed on the pavement and her husband simulated an accident.     

Fatal thoracic compression after being partially run over by an automobile

The authors report on a fatal pedestrian accident in which the victim lay on the street and got under a car being run over by one of the front wheels. Both the car driver and the killed pedestrian were strongly alcohol-intoxicated at the time of the accident (2.45 a.m.). After the car had come to a standstill, the pedestrian remained trapped under the car until she was rescued by the fire brigade. On the basis of the autopsy findings and the technical expert opinion it could be reconstructed that the pedestrian's death was not caused by the consequences of being run over by the car, but by the subsequent compression of her thorax in the final accident position.     

Fall from a Balcony—Accidental or Homicidal? Reconstruction by Numerical Simulation

In the case presented, conflicting witness accounts and the subject's injuries were highly suspicious of an assault that might have caused the balcony fall. For the reconstruction, a simulation software, originally designed for motor vehicle accident reconstruction, was used. Three scenarios were simulated using the PC‐Crash multibody pedestrian model: (S1) Subject was pushed against and fell over balcony rail, (S2) subject fell off from a seated position, (S3) subject fell off from a prone position on the rail. (S1) could be ruled out due to inconsistent results in terms of landing area and minimum velocity. Realistic results were obtained for (S3) with a fall off from a prone position on the rail. After a few months, the comatose subject awoke and gave an account of what had happened being consistent with the simulation results. This case demonstrates the feasibility of multibody simulations also in cases of nontraffic incidents.     

Two Simple Formulas for Evaluating the Lower Bound of the Impact Velocity in Vehicle–pedestrian Accidents

Formulas for evaluating the lower bound of the impact velocity are valuable in vehicle–pedestrian accident reconstruction. The theory of classical mechanics and four hypotheses were employed to derive formulas; the research results and simulation/accident tests were employed to validate their feasibility. Then, two simple formulas were developed according to the distance between the rest positions of the vehicle and the pedestrian and the flight‐phase distance. The results showed that the evaluated results by the two proposed formulas are inferior to the existing results. The influence of a roadside step on the impact velocity, which decreased with an increase in the flight‐phase distance and a reduction in the road slope, was evaluated. Based on a real accident, the study concludes that the lower bound can be easily obtained with the proposed formulas, which can be used to determine the evaluated impact velocity during simulations.     

Ankle joint injuries as a reconstruction parameter in car-to-pedestrian accidents

The study has assessed the usefulness of ankle joint injuries for deducing about the presumable location of a pedestrian in relation to a motor vehicle at the moment of collision. A statistically significant correlation was found between the direction of impact and the mechanism of ankle joint injuries. Such injuries were found in 113 of 317 pedestrian, fatal victims of road traffic accidents (36%). In the cases of impacts on the medial surface of the lower limb, the injuries associated with the supination mechanism were six times more frequent than the injuries associated with the pronation mechanism, while in the group of impacts on the limbs from the lateral side those connected with the pronation mechanism were four times more frequent. In victims hit from the back, the injuries connected with the plantarflexion mechanism were found almost four times more frequent than the dorsiflexion mechanism and in victims hit from the front in three of the five damaged joints the dorsiflexion component was observed (without any plantarflexion case). Less frequent "pronation injuries" in the limbs hit from the medial side and "supination injuries" in the limbs hit from the lateral side as well as "dorsiflexion injuries" in victims hit from the back predominantly occurred in impacts to the pelvic area, i.e. resulting from collisions with a van or lorry. Injuries of upper ankle joints are, thus, useful for reconstructing the presumable location of the pedestrian in relation to the vehicle at the moment of collision, especially, when the vehicle involved in the accident is known.     

Fatal pedestrian-bicycle collisions

Although, fatal collisions between pedestrians and bicycles are relatively rare, they are still of forensic relevance because of the need to explore the circumstances of the accident. Based on three reconstructed cases, situation and injury patterns are presented that might prove useful in future cases: usually the person causing the accident is the cyclist while the pedestrian generally suffers more severe injuries; the situation at the site of accident is important for its reconstruction: end location of the persons involved in the accident, injuries and traces on pedestrians and cyclists, traces at the site of accident and on the bicycle; because of the lack of pre-crash traces and any eyewitness accounts, the pedestrian's injuries are the best starting point for the reconstruction of the accident; a characteristic wound on the lower leg of the pedestrian that reveals the initial impact between the front wheel and the leg is crucial not because of its seriousness, but because of its external morphology; the injuries that can be expected by the following impact between body and handlebar are unspecific and only minor; the most severe injuries to the pedestrian as a result of the accident are caused secondarily by falling and hitting the head on the road; the fall of the cyclist, however, corresponds to a throw-off followed by a sliding phase with less impact load when the head hits the ground [maximum abbreviated injury scale 1 (MAIS 1)]; the cyclists involved are mainly younger persons on fashionable bicycles (here: mountain bikes); in the great majority of cases, the injured pedestrians are frail, elderly people with a lower tolerance of trauma.     

Clues in pedestrian-vehicle collisions and their significance for reconstruction of accidents

Investigations of clues in 87 fatal pedestrian-car accidents which took place with erect posture of the body are reported. Traces of paint proved to be most important clues transferred from the vehicle to the pedestrian's clothing. They derived from the outer paint layer of the collision sites at the front of the vehicle and occurred in 49% of the cases investigated. As has been shown by light and scanning electron microscopic investigations, there is intensive contact between the textile fibers and the paint traces which can be explained by action of heat in the collision. The higher the collision speed, the more readily paint is rubbed off. Because traces of rubbed off paint mark the site of the vehicle impact and indicate the color of the vehicle, they have a high evidence value for reconstruction of the accident. This also applies to traces of molten plastic material as well as to rubbed off chrome and rubber. In contrast to the prevalent view, paint splitters are unsuitable for determining the impact direction. Textile microtraces, traces of blood, hair and tissue occur as transferred clues from the pedestrian to the vehicle. These clues do not permit any statements with regard to the impact direction. They must be evaluated in connection with the remaining clues and the injuries.     

Improved Area of Origin Estimation for Bloodstain Pattern Analysis Using 3D Scanning

An important component of crime scene reconstruction is bloodstain pattern analysis (BPA). Where BPA concerns impact patterns, estimating the area of origin is critical information for scene reconstruction. Traditionally, this is achieved by measuring individual bloodstains and performing trigonometric calculations; however, 3D scanning has been proposed as a viable alternative for overcoming logistical and practical concerns with the manual method. Therefore, this project aimed to establish whether the FARO Focus 3D scanner and FARO Zone 3D software can improve the accuracy of area of origin estimates relative to the manual method. We created a series of eight bloodstain impact patterns and performed paired analysis using the two methods to estimate areas of origin for each pattern. Our data suggested that FARO-derived estimates were generally more accurate than using the manual method. FARO-estimated heights of origin areas were generally closer to the true distance. Both methods underestimated the distance from the wall for most patterns originating 150mm or greater from the wall, but overestimated distances for patterns originating closer to the wall. The degree to which distances were underestimated increased significantly the further the blood source was from the wall and was greater for FARO-derived estimates. The results of this research contribute to the validation of these instruments for operational implementation for BPA and should be considered alongside the practical benefits of 3D scanning relative to manual methods. Further, 3D scanning can provide reliable BPA reconstruction documentation for technical review and court presentation.     

Pedestrian accident analysis with a silicone dummy block

When a car is parked in an inclined plane in a parking lot, the car can roll down the slope and cause a pedestrian accident, even when the angle of inclination is small. A rolling car on a gentle slope seems to be easily halted by human power to prevent damage to the car or a possible accident. However, even if the car rolls down very slowly, it can cause severe injuries to a pedestrian, especially when the pedestrian cannot avoid the rolling car. In an accident case that happened in our province, a pedestrian was injured by a rolling car, which had been parked on a slope the night before. The accident occurred in the parking lot of an apartment complex. The parking lot seemed almost flat with the naked eye. We conducted a rolling test with the accident vehicle at the site. The car was made to roll down the slope by purely gravitational pull and was made to collide with the silicone block leaning against the retaining wall. Silicone has characteristics similar to those of a human body, especially with respect to stiffness. In the experiment, we measured the shock power quantitatively. The results showed that a rolling car could severely damage the chest of a pedestrian, even if it moved very slowly.     

Correlations of injury, toxicology, and cause of death to Galaxy Flight 203 crash site

This report details the medical investigation of the crash of Galaxy Flight 203, on 21 Jan. 1985, near Reno, Nevada. Sixty-eight persons died at the scene, two died during hospitalization, and one victim survived. After completion of autopsies on all victims, pathologists determined causes of death based upon injuries, evidence of smoke inhalation, and toxicologic results. Our research shows that the majority of victims survived the impact only to succumb to toxic gas and fire. We correlated the causes of death, various injuries, and toxicologic findings to body location at the crash site to aid in reconstructing the events surrounding the accident and to address medicolegal problems and safety considerations. Our experience clearly supports the need for thorough medical investigation, including autopsy of each victim and determination of the precise cause of death for all fatalities.     

Artifacts Caused by Leaf-Cutting Ants of the Genus Atta (Hymenoptera: Formicidae): Postmortem Bite Injuries and the Tearing of Clothes

Ants are one of the first insects to find an exposed cadaver and can be present during all stages of decomposition. Although these organisms are not commonly used in postmortem interval estimates, they are to be taken into account on criminal investigations involving human corpses, since they can leave bite marks that can be mistaken for antemortem or perimortem injuries, which could be misleading when ascertaining the occurrence of abuse or physical altercation during a crime. A few studies report the action of ants on human cadavers and even though leaf-cutting ants of the genus Atta are frequently encountered in succession studies that use animal carcasses, there are no records of these fungus-growing species on human corpses. Atta is a genus restricted to the New World, ranging from northern Argentina to southern United States and acts as one of the most conspicuous neotropical herbivores. In this study, we report three cases of violent death that illustrate the impact of ants, especially those of the genus Atta, in a forensic setting. We compare the patterns displayed by postmortem bite injuries caused by leaf-cutter ants and other common species with less robust mandibles. We also present the capability of Atta ants to create artifacts by cutting victim’s clothes in a crime scene, contributing to the knowledge of ant-mediated confounding factors in crime scene investigation.     

The Reconstruction of the Dynamic of a Murder Using 3D Motion Capture and 3D Model Buildings: The Investigation of a Dubious Forensic Case

Photographic surveying is fundamental to crime scene investigation. It is performed using 2D images; however, this method is limited because it does not offer a detailed view. We present a 3D (three‐dimensional) experimental model that has clarified unsolved aspects of a homicide. A woman was found dead in her home. Considering the large number of injuries on the victim, signs of the use of two weapons, and the absence of traces of the killer/s, several hypotheses about the dynamics and the number of offenders were made. The forensic surveys suggested that a single offender had used two weapons at the same time, but this hypothesis was considered to be impossible by investigators. Our 3D model allowed us to experimentally reproduce the murder and compare various dynamics. The model was used as evidence in Court to prove the hypothesis of a single killer and demonstrate that no other people were involved.