Evan S Walsh

A centric/non-centric impact protocol and finite element model methodology for the evaluation of American football helmets to evaluate risk of concussion

American football reports high incidences of head injuries, in particular, concussion. Research has described concussion as primarily a rotation dominant injury affecting the diffuse areas of brain tissue. Current standards do not measure how helmets manage rotational acceleration or how acceleration loading curves influence brain deformation from an impact and thus are missing important information in terms of how concussions occur. The purpose of this study was to investigate a proposed three-dimensional impact protocol for use in evaluating football helmets. The dynamic responses resulting from centric and non-centric impact conditions were examined to ascertain the influence they have on brain deformations in different functional regions of the brain that are linked to concussive symptoms. A centric and non-centric protocol was used to impact an American football helmet; the resulting dynamic response data was used in conjunction with a three-dimensional finite element analysis of the human brain to calculate brain tissue deformation. The direction of impact created unique loading conditions, resulting in peaks in different regions of the brain associated with concussive symptoms. The linear and rotational accelerations were not predictive of the brain deformation metrics used in this study. In conclusion, the test protocol used in this study revealed that impact conditions influences the region of loading in functional regions of brain tissue that are associated with the symptoms of concussion. The protocol also demonstrated that using brain deformation metrics may be more appropriate when evaluating risk of concussion than using dynamic response data alone.

Comparison between Hybrid III and Hodgson–WSU headforms by linear and angular dynamic impact response

In brain injury research, linear and angular resultant acceleration data have been considered important mechanisms contributing to various levels of brain injury. The development of biofidelic headforms with similar dimensions and weight to that of a real human head has allowed for researchers to repeatedly collect data related to the effects of different impacts on the human head. Currently, there are different types of headforms available for impact testing, each with varying degrees of biofidelity and repeatability. Two commonly used headforms were tested: the Hybrid III and the Hodgson–WSU (NOCSAE). The two headforms were outfitted with nine single-axis accelerometers positioned orthogonally following a 3–2–2–2 array. Both headforms show good linearity and correlate well throughout the different velocities and are, therefore, reliable tools. Significant differences are observed in peak linear and peak angular accelerations between Hybrid III and Hodgson–WSU headforms. The shapes of the loading curves are visually different and thus may have significant impact on the output from FE modelling of the brain response.

Dynamic impact response characteristics of a helmeted Hybrid III headform using a centric and non-centric impact protocol

A linear impactor system was used to apply a condensed version of the University of Ottawa Test Protocol, employing five centric and non-centric impact conditions, to a Hybrid III headform fitted with six certified ice hockey helmets. None of the helmeted conditions exceeded linear acceleration thresholds for traumatic or mild traumatic brain injury; however, five of the six helmets had angular acceleration results that were above the 80% risk of mild traumatic brain injury threshold proposed by Zhang et al. High risk of mild traumatic brain injury was associated with non-centric impact conditions and peak angular accelerations, supporting the need for improved three-dimensional helmet certification standards.

Analysis of loading curve characteristics on the production of brain deformation metrics

Traumatic and mild traumatic brain injuries are incurred as a result of the complex motions of the head after an impact. These motions can be quantified in terms of linear and rotational accelerations which cause the injurious levels of brain deformation. Currently, it is unclear what aspects of the linear and rotational acceleration loading curves influence injurious brain deformation. This research uses the University College Dublin Brain Trauma Model to analyse the loading curve shapes from a series of centric and non-centric impacts to a Hybrid III headform fitted with different hockey helmets. The results found that peak resultant linear acceleration did not always correlate with brain deformation measures. The results also indicated that, due to the complex nature of the interaction between loading curve characteristic and tissue parameters, there was no commonality in curve shape which produced large magnitudes of brain deformation. However, the discriminant function did show that angular acceleration loading curve characteristics would predict brain deformation more reliably than linear acceleration loading curves.

The application of brain tissue deformation values in assessing the safety performance of ice hockey helmets

This research was undertaken to examine a new method for assessing the performance of ice hockey helmets. It has been proposed that the current centric impact standards for ice hockey helmets, measuring peak linear acceleration, have effectively eliminated traumatic head injuries in the sport, but that angular acceleration and brain tissue deformation metrics are more sensitive to the conditions associated with concussive injuries, which continue to be a common injury. Ice hockey helmets were impacted using both centric and non-centric impact protocols at 7.5 m/s using a linear impactor. Dynamic impact responses and brain tissue deformations from the helmeted centric and non-centric head form impacts were assessed with respect to proposed concussive injury thresholds from the literature. The results of the helmet impacts showed that the method used was sensitive enough to distinguish differences in performance between helmet models. The results have shown that peak linear acceleration yielded low magnitudes of response to an impact, but peak angular acceleration and brain deformation metrics consistently reported higher magnitudes, reflecting a high risk for incurring a mild traumatic brain injury.

Determination of high-risk impact sites on a Hybrid III headform by finite element analysis

The current standards and methods to evaluate helmet performance remain focused on traumatic brain injuries and not concussive injuries. This is reflected in the methodologies currently used and the injury metrics employed to determine the pass/fail criteria for the helmets being tested. To address the problem surrounding concussion and helmets, a method reflecting high risk of concussive injury must be developed. The purpose of this research was to identify high risk of concussive injury impact sites on the Hybrid III headform using the Wayne State head injury finite element model. The Hybrid III headform was impacted using a linear impactor in five different sites with four angles per site at 5.5 m/s. The resulting acceleration loading curves were used as input for brain deformation analyses using the Wayne State University Brain Injury Model. The brain deformation results indicated that there are 12 impact conditions on the Hybrid III headform which reflect a risk of concussion above 80% when compared with the literature. The method used and impact sites discovered by this research differ significantly from the current standard method and may be used to guide future impact sites for the evaluation of helmet performance.

Evaluation of Dynamic Response and Brain Deformation Metrics for a Helmeted and Non-Helmeted Hybrid III Headform Using a Monorail Centric/Non-Centric Protocol

Head injuries and concussion in particular has become a source of interest in the sport of ice hockey. This study proposes a monorail test methodology combined with a finite element method to evaluate ice hockey helmets in a centric/non-centric protocol with performance metrics more closely associated with risk of concussion. Two conditions were tested using the protocol a) helmeted vs no helmet, and b) vinyl nitrile lined hockey helmet vs expanded polypropylene lined hockey helmet. Results indicated that the impact velocities and locations produced distinct responses. Also, the protocol distinguished important design characteristics between the two helmet liner types with the vinyl nitrile lined helmet producing lower strain responses in the cerebrum. Furthermore, it was discovered that low risk of injury peak linear and rotational acceleration values can combine to produce much higher risks of injury when using brain deformation metrics. In conclusion, the use of finite element modeling of the human brain along with a centric/non-centric protocol provides an opportunity for researchers and helmet developers to observe how the dynamic response produced from these impacts influence brain tissue deformation and injury risk. This type of centric/non centric physical to finite element modeling methodology could be used to guide innovation for new methods to prevent concussion. Keywords: Ice hockey, Helmets, Standards, Concussion 1.0 Introduction Mandatory protective headgear in impact and contact sports help protect athletes against traumatic brain injuries (TBI) including intracranial bleeds and skull fractures. However, mild traumatic brain injuries (mTBI), such as concussions, are still common with studies reporting helmets not effective in managing the risk of mTBI [1,2]. The National Hockey League (NHL) report an increase in mTBIs’ over the last decade accounting for 18% of all hockey injuries [4]. These statistics suggest that changes in the game including improved helmet technology have had little effect on the incidence of concussion. Present helmet technology is designed to minimize peak linear acceleration during a direct impact [6]. Linear acceleration was chosen as the performance metric for evaluating helmets as this measure has been associated with TBI [6;7;8]. As a result, linear dominant impact conditions have been utilized in standards to evaluate helmets [9;10]. These standards typically use a headform and monorail system for primarily centric (defined as the impact vector passing through the center of gravity of the head) impacts. However, rotational acceleration has also been identified as an important factor in the incidence of concussion and must also be measured. Higher rotational acceleration responses tend to result from non-centric impacts (defined as impacts whose vector does not pass through the centre of gravity of the head) [11;12]. These rotations cause shear stress within the brain which has been proposed as a predictor for mTBI [11;12]. Current helmet standards do not consider rotational acceleration when assessing helmet performance despite several studies associating rotational acceleration to risk of sustaining a concussion [5;12;13]. However, definitive thresholds of injury for concussion using linear and rotational acceleration have yet to be elucidated; this difficulty has been identified by researchers to be due to the kinematics not accounting for the interaction between the impact induced motions and the brain tissue [15;16]. As a result, advanced computational models have been developed to better understand the effect of impact head kinematics on brain tissue damage [13;14;15]. Measuring brain tissue deformation using finite element models of the human brain is considered an effective method in evaluating risk of sustaining an mTBI [16]. Finite element modeling of the brain during impact allows for the examination of the effect of complex loading curves on brain tissue deformations. The characteristics of these linear and rotational acceleration loading curves can then be used as input parameters into complex brain models which can then simulate the deformation of tissue resulting from the kinematics of an impact event [17;18]. Past research has shown how this method can predict the effect of linear and rotational accelerations on the stresses and strains imparted to the brain through car crash analysis as well as hockey and football helmet impacts [13;19;20]. As a result, finite element models for the head and brain provide an opportunity to use brain deformation values to evaluate the ability of a hockey helmet to reduce the risk of brain injury [20]. There is presently no standard which uses a centric/non-centric impact method coupled with finite element analysis to measure brain deformations from helmeted impacts. If such a method was developed it may aid in supplying more information on helmet performance using linear and rotational acceleration as well as brain deformation metrics [13;14;15;21]. Previous research has investigated this type of protocol using a linear impactor system, which was created to replicate player to player collisions [5]. This linear impactor method is different from current drop tower methods used by certification bodies to certify helmets. The development of this type of protocol using the monorail drop system to include centric and non-centric impacts may allow for easier adoption this new protocol using current test equipment. The objective of this study was to use a monorail centric/non-centric impact methodology to compare the dynamic responses of a helmeted and un-helmeted Hybrid III headform. In addition, VN and EPP helmets were tested determine if there is any difference in the management of linear and rotational acceleration between these impact absorbing liners using the proposed protocol. 2.0 Methodology 2.1 Equipment A monorail drop rig was used (Figure 1) to complete the proposed testing protocol for the evaluation of the performance of hockey helmets. For the purpose of this study a 50 percentile male Hybrid III headand neckform (mass 6.08kg ± 0.01kg) was attached to the drop carriage by the base of the neckform with a special jig designed to ensure a 90° angle between the z-axis of the headform and the monorail (Figure 2). A 0.46 ± 0.01m tall anvil extension 0.104 ± 0.05m in diameter was firmly fixed to the monorail base. For non-centric impacts the anvil extension was moved horizontally 6.5cm in line with the x-axis of the headform and secured with C-clamps. Secured on the tip of the impact anvil was a hemispherical nylon pad (diameter 0.126 ± 0.01m) covering a modular elastomer programmer (MEP) 60 Shore Type A (0.025 ± 0.05m thickness) disc (Figure 3). Together the nylon pad and MEP disc weighed 0.908 ± 0.001kg. The MEP was chosen as it is a common material used in helmet standards (CSA; NOCSAE). The nylon and MEP disc combination was not designed to reflect any particular impact scenario on the ice. A 50 percentile adult male Hybrid III headform (mass 4.54 kg ± 0.01kg) (Figure 4) was used in this study. This type of headform is designed to respond in a reproducible and reliable manner and is primarily used in impact reconstructions [22]. The headform was instrumented with nine single-axis Endevco7264C-2KTZ-2-300 accelerometers according to Padgaonkar’s orthogonal 3-2-2-2 linear accelerometer array protocol to measure the three dimensional kinematics of the head from an impact [23]. The headform coordinate system was defined with a left-hand rule. Positive axes were directed toward the anterior, toward the right ear and caudally for x, y and z respectively. The Hybrid III neck with a mass of 1.54 ± 0.05 kg was composed of 4 butyl rubber discs interlocked between five aluminum plates to simulate human vertebrae. The discs were offset towards the front 0.5cm and were slit to elicit a different response in flexion from that in extension [24]. 2.2 Data Collection Inbound velocity was set using the Cadex Impact v5.7a computer program and recorded using a velocimeter (time gate). The nine mounted single-axis Endevco7264C-2KTZ-2-300 accelerometers (Endevco, San Juan Capistrano, CA) were sampled at 20 kHz and the signals were passed through a TDAS Pro Lab system (DTS, Calabasas, CA) prior to being processed by TDAS software. 2.3 Procedure The Hybrid III was dropped at three different inbound velocities (2, 4 & 6 m/s) in order to examine how the dynamic response changes as velocity increased (Marino and Drouin, 2000). Three impact conditions were chosen for preliminary investigation of non-centric impacts using the monorail drop rig and are shown/listed in Figure 5 and Table 1 [25]. Two models of helmets were tested, with three helmets of each model used for a total of 6 helmets impacted. Each model had identical two piece polyethylene (2PE) shells with either VN or EPP liners. Dimensions of the shell and foam liner are described in Table 2. The headform and helmeted headform was impacted using a monorail drop rig and each condition tested three consecutive times, which is standard procedure for testing multiple-impact helmets [9;10]. During testing the average time between impacts was 5 ± 0.50 min, which exceeds requirements by current standards [9;10]. Impact site accuracy was ensured by marking the helmet with a permanent marker when it was in contact with the impact cap prior to the first drop. The helmet was reset after each impact to ensure the mark on the helmet was in line with the mark on the impact cap. A different helmet was used for each impact velocity; therefore a total of 162 total helmeted impacts were performed. For the un-helmeted headform condition there was a total of 81 impacts. 2.4 Finite element model (University College Dublin Brain Trauma Model) In addition, the resulting three-dimensional loading curve responses (x, y and z) were applied to the University College Dublin (UCDBTM) finite element model to pro