Falls and collisions are the most common types of events leading to sports-related concussions where impacts to the head play an important role on the onset of traumatic brain injury. Each event can be described by impact parameters that define the loading conditions on the head and brain and are necessary for accurate accident reconstruction employing physical impact tests, anthropometric headforms, and finite element (FE) modelling. It was the purpose of this research to describe the effects and interactions of impact velocity, compliance, mass and impact location on head acceleration and brain tissue strain measures associated with risk of concussions in sports. Impact parameters were varied to capture responses from no-injury up to concussive levels. Study one examined the effect of impact parameters on fall events simulated using a monorail drop tower. Impact mass was varied using three different headforms representing child, adolescent, and adult sizes measuring peak linear and angular acceleration and maximum principal strain. Regression analysis revealed that impact compliance was the most influential on peak linear and angular acceleration measures, meanwhile FE strain was most affected by changes in impact velocity. Smaller headforms tend to produce higher acceleration and strain values, supporting the need for age and size related mechanical definitions of risk. Study two examined the effect of impact parameters for collision events simulated using a multi-mass pendulum to represent common striking masses in sport measuring peak linear and angular acceleration and strain. Study three provided further insight into collision impacts by evaluating the distribution of peak strains in different brain lobes and the volume of the brain experiencing strains passed a critical level. Results show that compliance was similarly the most influential on peak head acceleration whereas peak strain and volume were most affected by impact velocity. Mass-velocity interactions had effects where a 9 kg mass had greater response than 15 kg, but similar to 21 kg. The temporal lobe consistently contained the highest strains with the rear boss non-centric impact location producing the largest values. Interacting impact parameters illustrate the challenges with predicting associated risk of concussion from head collisions in sport and supports the need to identify effective performance ranges of protective materials.

The objective of this research was to better understand how impact angle influences headform dynamic response and brain tissue deformation. A bare headform was impacted using a pneumatic linear impactor at 5.5 m/s. The impacts were directed on the front and side location at angles of 0, 5, 10 and 15° rightward rotations as well as -5, -10 and -15° (leftward) rotations at the side to examine the characteristics of the head and neckform on the results. Peak resultant linear and rotational accelerations from the headform as well as peak maximum principal strain (MPS) and von Mises stress (VMS) estimated from a brain finite element model were used to measure the effect of impact angle. Significant results were dependent upon the impact angle and location as well as the dependent variable used for comparison (p <0.05). Impact angle produced significant differences in rotational acceleration and MPS at both the front and side; however angle only had an effect on VMS and linear acceleration at the front and side locations, respectively. These findings show that the effect of impact angle is asymmetrical and is specific to the dependent variable. This study suggests that varying impact angle alone may not be as influential on headform dynamic response and brain tissue deformation and that the severity of an impact may be more of a function of how both location and angle create high risk conditions.

The objective of this research was to better understand how impact angle influences headform dynamic response and brain tissue deformation. A bare headform was impacted using a pneumatic linear impactor at 5.5 m/s. The impacts were directed on the front and side location at angles of 0, 5, 10 and 15° rightward rotations as well as -5, -10 and -15° (leftward) rotations at the side to examine the characteristics of the head and neckform on the results. Peak resultant linear and rotational accelerations from the headform as well as peak maximum principal strain (MPS) and von Mises stress (VMS) estimated from a brain finite element model were used to measure the effect of impact angle. Significant results were dependent upon the impact angle and location as well as the dependent variable used for comparison (p <0.05). Impact angle produced significant differences in rotational acceleration and MPS at both the front and side; however angle only had an effect on VMS and linear acceleration at the front and side locations, respectively. These findings show that the effect of impact angle is asymmetrical and is specific to the dependent variable. This study suggests that varying impact angle alone may not be as influential on headform dynamic response and brain tissue deformation and that the severity of an impact may be more of a function of how both location and angle create high risk conditions.

The objective of this research was to better understand how impact angle influences headform dynamic response and brain tissue deformation. A bare headform was impacted using a pneumatic linear impactor at 5.5 m/s. The impacts were directed on the front and side location at angles of 0, 5, 10 and 15° rightward rotations as well as -5, -10 and -15° (leftward) rotations at the side to examine the characteristics of the head and neckform on the results. Peak resultant linear and rotational accelerations from the headform as well as peak maximum principal strain (MPS) and von Mises stress (VMS) estimated from a brain finite element model were used to measure the effect of impact angle. Significant results were dependent upon the impact angle and location as well as the dependent variable used for comparison (p <0.05). Impact angle produced significant differences in rotational acceleration and MPS at both the front and side; however angle only had an effect on VMS and linear acceleration at the front and side locations, respectively. These findings show that the effect of impact angle is asymmetrical and is specific to the dependent variable. This study suggests that varying impact angle alone may not be as influential on headform dynamic response and brain tissue deformation and that the severity of an impact may be more of a function of how both location and angle create high risk conditions.

The incidence of concussive injury has continued to arise annually with up to 3.8 million concussions reported per year (Thurman 1999) and 15% of these injuries occurring with persistent symptoms (Kraus and Chu, 2005). Few studies have examined the differences between youth and adult concussion (Yeates et al, 2012; Gosselin et al, 2010) therefore it is unknown whether youth and adults pose a similar risk of sustaining a concussion following impact. For this reason, the purpose of this study is to determine if differences exist in the dynamic response of the head and brain tissue deformation characteristics between children and adolescents for falls in comparison to adult data which have resulted in concussive injuries. Patient data was collected from emergency room hospitals across Canada. After exclusion criterion was applied, 11 child and 10 adolescent falls were reconstructed using mathematical (MADYMO) model, physical model (Hybrid III Headforms) and finite element modelling. Both groups were compared to each other as well as an adult group collected by Post et al (2014b) using a one-way ANOVA and Welsh test. The results of this study show that the children produced the lowest values for all variables when compared to the adolescents and adults whereas the adolescents produced the largest (with the exception of MPS where the adolescent and adult MPS was the same). Although all results were above the suggested thresholds for risk of concussive injury, the youth produced the lowest brain tissue strain yet still suffered a concussion. This is important to note as it may suggest that children are at an increased risk of injury at a lower brain tissue strain level. Understanding the differences in parameters influencing concussive injury may aid researchers in comprehending the unique risk for youth at difference ages. This information would be useful in terms of protective equipment design, promoting safe play in games and management of patients following injury.

Frontal motor vehicle collisions (MVCs) account for the majority of injuries and fatalities in MVCs according to the Fatality Analysis Reporting Systems (FARS). One of the most commonly injured regions of the body during MVCs is the thorax. While there are fewer adult passengers riding in the rear seat compared to the front seat, the number of adults in the rear seat may increase dramatically in the near future with the rise of ridesharing services and highly automated vehicles (HAVs). With the increase in exposure for adults riding in the rear seat, the safety of these passengers needs to be evaluated. Previous research has shown that occupant protection in the rear seat is disproportionately lower than that of the front seat in modern vehicles due to the focus on front seat occupants in both regulatory and market-driven crash tests. This has resulted in many of the occupant safety systems, e.g., pretensioners (PT), load limiters (LL), and airbags, being widely available in the front seat, but sparsely available in the rear seat. Anthropomorphic test devices (ATDs) have been developed to investigate occupant safety during frontal MVCs and can be utilized in the investigation of rear seat occupant injuries. However, the biofidelity and injury risk criteria used for these ATDs has only been validated when seated in the front seat. To validate the response and injury risk predictions of existing frontal ATDs in the rear seat it is necessary to generate new biomechanical data in the rear seat of modern vehicles. The purpose of this work is to quantify the biomechanical responses of two frontal ATDs, i.e., the Hybrid III and THOR-50M 50th percentile male ATDs, and 50th percentile male post mortem human surrogates (PMHS) seated in the rear seat of modern vehicles, which have various seat geometries and restraint types, during frontal MVCs. Emphasis is placed on comparisons between the thoracic responses of the three human surrogates e.g., thoracic deflection time histories, and thoracic injury risks, i.e., ATD injury risk prediction versus instances of PMHS injuries. A series of twenty-four frontal sled tests were first conducted with the HIII and THOR-50M ATDs seated in the rear seats of seven vehicle test bucks with varying seat geometries and two different restraint types. Three vehicles had advanced restraints while four had conventional restraints. Three different crash pulses were used derived from vehicle specific US New Car Assessment Program frontal crash data: Scaled (32kph), Generic (32kph), and NCAP85 (56kph). Thoracic injury metrics were not exceeded in the lower severity pulses for either ATD but were exceeded during some of the high severity tests. A matched comparison analysis between a front and rear seated Hybrid III 50th percentile male ATD is presented second that highlights the disparities between front and rear seat iii occupant safety of modern vehicles during frontal MVCs. The Hybrid III ATD data were used for this comparison. Thoracic injury risk was found to be higher for the rear seated HIII across all vehicles, while thoracic acceleration was lower in the rear than the front for some vehicles. PMHS thoracic responses and injury risk equations were then evaluated in four of the vehicles used for the ATD tests using the high severity sled pulse, i.e., NCAP85 (56kph). Thoracic acceleration and normalized deflection values were higher in vehicles with conventional restraints, and the location of maximum deflection was always inboard of the sternum. The resulting thoracic injuries ranged from AIS 3 to AIS 5. Additionally, there were a larger average number of rib fractures in vehicles with conventional restraints versus advanced restraints. A multi-point deflection injury risk equation predicted injury the best. However the less censored rib fracture data that were obtained suggest that all three of the injury equations evaluated could be improved. Lastly, the PMHS data were used to assess the similarities in thoracic response between the ATDs and PMHS. An objective rating metric was used for the response comparison. The HIII had a slightly better average score than the THOR-50M; however, the THOR-50M had a more biofidelic kinematic response during the tests. This analysis furthers the understanding of the effect of different occupant protection systems on thoracic injury risk in a rear seat environment and the biofidelity of frontal 50th percentile male ATDs in the rear seat. / Doctor of Philosophy / Frontal motor vehicle collisions (MVCs) account for the majority of injuries and fatalities in MVCs according to the Fatality Analysis Reporting Systems (FARS), a nationwide census of fatal injuries suffered during crashes. One of the most commonly injured regions of the body during MVCs is the thorax i.e. the chest. While there are fewer adult passengers riding in the rear seat compared to the front seat, the number of adults in the rear seat may increase dramatically in the near future with the rise of ridesharing services and in the future, the rise of highly automated vehicles (HAVs commonly called "driverless cars"). The safety of adult rear seat passengers needs to be evaluated due to the potential increase in occupancy rates. Previous research has shown that occupant protection in the rear seat is disproportionately lower than that of the front seat in modern vehicles. This is likely due to the focus on front seat occupants in both regulatory tests and market-driven crash tests such as the New Car Assessment Program and IIHS frontal overlap tests. This has resulted in many of the advanced occupant protection systems being widely available in the front seat, but sparsely available in the rear seat. Anthropomorphic test devices (ATDs), i.e., crash test dummies, have been developed to investigate occupant safety during frontal MVCs and can be utilized in the investigation of rear seat occupant injuries. However, the biofidelity (similarity of ATD response to a human surrogate) and injury risk criteria used for these ATDs has only been validated when seated in the front seat. To validate the thoracic response and injury risk predictions of the existing frontal ATDs when seated in the rear seat it is necessary to generate new biomechanical data in the rear seat of modern vehicles. The purpose of this work is to quantify the thoracic response of two current 50th percentile male frontal impact ATDs, i.e., the Hybrid III and THOR-50M, and similarly sized male post mortem human surrogates (PMHS) seated in the rear seat during a frontal MVC. Several vehicles were used and chosen to represent various seat geometries and restraint types. There are two restraint types in the rear seat within this body of work, conventional and advanced. A conventional restraint consists of a three point seat belt, while an advanced restraint consists of a three point seat belt with additional safety features installed. Emphasis is placed on the injury risk prediction from the ATD versus actual instances of injuries from the PMHS. A series of frontal sled tests were first performed with the Hybrid III and THOR-50M ATDs. Three different crash pulses derived from vehicle specific US New Car Assessment Program frontal crash data were used: Scaled (32kph), Generic (32kph), and NCAP85 (56kph). v These tests showed that the established injury metrics for the two ATDs were exceeded in some of the high severity tests. A matched comparison analysis between a front and rear seated Hybrid III 50th percentile male ATD is presented and highlights the disparities between front and rear seat occupant safety of modern vehicles during frontal MVCs. The thoracic injury risk was found to be higher in the rear compared to the front for all vehicles. A series of frontal sled tests were then performed with the mid-sized male PMHS using the high severity sled pulse (NCAP85) and four of the vehicles from the ATD tests. The thoracic deflections for the PMHS were normalized by the surrogate chest depth in order to compare them between different sized surrogates, and were found to be higher in vehicles with conventional restraints. All PMHS had severity thoracic injuries. Additionally, there were a larger average number of rib fractures in vehicles with conventional restraints versus advanced restraints. Finally, the thoracic response of each ATD was compared to the PMHS to further the understanding of the effect of different occupant protection systems on thoracic injury risk in a rear seat environment and investigate rear seat biofidelity of each ATD. The THOR-50M had a more biofidelic kinematic response, while the Hybrid III matched the PMHS thoracic deflections and accelerations more accurately when compared with an objective rating metric. The comparison between surrogate responses furthers the understanding of 50th percentile male ATD biofidelity, the ATD injury risk prediction capabilities, and effects of different occupant protection systems on thoracic injuries in the rear seat.

No description available.

Whiplash injuries as a result of rear impact are among the most common injuries in traffic accidents. This is why whiplash injuries have reached a high priority in the research area. In this study, the effects of head restraint position and impact pulse to the whiplash injury have been analyzed by performing whiplash tests using the sled test facility of METU-BILTIR Center Vehicle Safety Unit. Although there are many whiplash test protocols, the test sample has been prepared according to the most recent one, Euro NCAP Whiplash Test Protocol. Three different head restraint positions and three different impact pulses with different severities, totally nine tests have been performed. The tests are performed with a three point generic seat belt and an instrumented Hybrid III 50th percentile male adult crash test dummy is used as the occupant in driver seat of a light commercial vehicle. High speed cameras, sensors on the crash test dummy and a data acquisition system are used to take the test data. This test data has been analyzed and presented according to the defined whiplash assessment criteria and the performance scores of the particular seat for each test have been determined using the whiplash assessment criteria values according to the Euro NCAP Test Protocols.

Avec le développement de l’usage d’internet et les nouveaux services tout en ligne, la quantité de données traitée par les data-centers ne cessent de croître. Ainsi, si la mise en parallèle de plusieurs serveurs permet de répondre à cette demande, un problème structurel apparaît. Comme dans les supercalculateurs entre les noeuds de calculs, les données ne circulent plus suffisamment rapidement entre les serveurs sur les câbles électriques classiques. Pour pallier à ce goulot d’étranglement, l’utilisation de l’optique permet d’obtenir des débits plus importants. Si les câbles actifs existants permettent une solution rapide, la photonique sur silicium présente un avantage certain. L’intégration des composants optiques au plus près des puces électroniques permet de réduire considérablement le chemin des interconnexions ainsi que leurs coûts énergétiques. Une chaine de communication optique complète nécessite différents composants. Si les modulateurs, multiplexeurs, coupleurs fibres, démultiplexeurs et photodetecteurs ont déjà été démontrés, les sources lasers utilisées sont toujours extérieures à la puce photonique. Il s’agit en effet du chainon manquant dans l’intégration complète de l’optique grâce à la photonique sur silicium. Plusieurs architectures ont déjà été proposées mais cette thèse s’appuie sur l’intégration hybride d’un matériau III-V sur le silicium. Le travail de cette thèse a consisté en la conception, la fabrication et la caractérisation de sources laser hybrides III-V sur silicium et a été entièrement accompli aux CEA/LETI. L’architecture du LETI d’un guide III-V couplé à un guide silicium a été améliorée grâce à un critère adiabatique pour obtenir une zone active de laser efficace et robuste. Cette architecture a été déclinée en différents types de lasers (Fabry-Pérot, DBR, racetrack et DFB). La fabrication de ces lasers a nécessité des développements de procédés de structuration du matériau III-V reporté sur du silicium dans les laboratoires du CEA/LETI. Les premiers résultats ont permis la validation de l’architecture utilisée. Les lasers DBR présentent des seuils inférieurs à 20mA et des puissances optiques maximales supérieures à 20mW dans le guide silicium. Ces lasers ont également un fonctionnement monomode avec un SMSR de plus de 50dB. Les lasers DFB possèdent quant à eux des seuils de 30mA et des puissances optiques maximales supérieurs à 40mW dans le guide silicium. Ils sont monomodes avec un SMSR de 40dB. Ces résultats sont à l’état de l’art mondial sur les sources lasers hybrides en photonique sur silicium. / With the development of the Internet and the new cloud services, the amount of data processed by data-centers is increasing. Though, if the paralleling of multiple server answer to this growing quantity, a structural problem arises. As in super calculators between nodes calculations, data are not transmitted quickly enough between servers on classical electric cables. This bottleneck can be overcome thanks to the optic which can access greater data rates. If existing active cables allow a quick resolution, silicon photonic has a clear benefit. The integration of the optical components closer to the electronic chips reduces substantially the path of interconnections and their energetic costs.An optical transmitter and receptor need different components. If modulators, multiplexers, fiber coupler, multiplexer and photo-detectors are already achieved, laser sources used are still outside the photonic chip. This is the missing link for a complete optical integration thanks to the silicon photonic. Several architectures have been proposed but this thesis relies on hybrid integration of III-V material on silicon.The work of this thesis consisted on the conception, the fabrication and the characterization of hybrid III-V on Silicon laser sources and was completely done at the CEA/LETI. The LETI architecture composed by a III-V waveguide coupled to a silicon waveguide was improved thanks to a adiabatic criterion to obtain an efficient and robust active area of the laser. This architecture was declined in different kinds of lasers (Fabry-Pérot, DBR, Racetrack and DFB). The fabrication required technological development for the structuration of the reported III-V material on silicon at the laboratories of the CEA/LETI. The first results validates the proposed architectures. The DBR lasers have threshold of less than 20mA and maximal optical power of more than 20mW inside the silicon waveguide. Those lasers are monomode with a SMSR of more than 50dB. The DFB Lasers have threshold of 30mA and maximal optical power of more than 40mW inside the silicon waveguide. They are monomode with SMSR more than 40dB. Those results are world state-of-the-art for hybrids laser sources in silicon photonic.

No description available.