Football helmets were originally introduced in 1896 as protective sports equipment, with devices fashioned out of leather strips. The evolution of leather caps involved the integration of a suspension system, as in the 1930s MacGregor-Goldsmith leather helmet. However, when helmets were introduced, the incidence of head injuries actually increased because of the pervasiveness of head-to-head collisions.
In response to an epidemic of football-related fatalities during the 1939 season, the National Collegiate Athletic Association (NCAA) mandated helmet use at the collegiate football level. The National Football League (NFL) followed suit for professional players in 1943. Football helmet design evolved significantly around this period, with the development of the first plastic shell helmet in 1937.
After 38 fatalities on the football field in 1969, the National Operating Committee on Standards for Athletic Equipment (NOCSAE) was formed to establish standards for helmet testing. Specifications were revised in 1977 to include procedures for periodic helmet recertification, which were adopted by the NCAA and the National Federation of State High School Associations in 1978, mandating that all players wear helmets that meet NOCSAE test standards.
In the early 1970s, the American Society for Testing Materials (ASTM) formed a committee (F-08) to develop a test method for shock attenuation in football helmets.2 While the ASTM defines pass/fail criteria for football helmets based purely on linear accelerations (currently 275-g threshold), the NOCSAE test method uses the Gadd Severity Index (GSI),7 an acceleration-time tolerance profile based on fracture patterns of animal and human cadaveric skulls. The NOCSAE uses a twin wire-guided drop test whereby a helmeted surrogate head, without the neck, is impacted onto an anvil. The pass/fail threshold for football helmets is based on a GSI score of 1200. None of the current standards for assessing the performance of football helmets, or helmets for any sports application in the US, require measurement of the types of forces believed to be responsible for rotational brain injuries, including concussion and subdural hematoma.
According to data from the National Center for Catastrophic Sports Injury Research, 243 football deaths were recorded between 1990 and 2010 among high school and college football players.3 This is equal to an average of 12 deaths each year in the US. Sixty-two of these deaths were attributed to brain injury, typically subdural hematoma. Concussion is considerably more prevalent. In the US it is estimated that 300,000 sports-related concussions occur annually. Among individuals 15–24 years of age, sports are second only to motor vehicle crashes as the leading cause of concussion. Football is reported to have the highest incidence of sports-related concussion, followed by hockey and lacrosse.18
Biomechanics of Brain Injury
Traumatic brain injury is associated with 2 primary mechanisms: impact loading and impulse loading. A direct blow transmitted primarily through the center of mass of the head produces impact loading, which can result in extracranial focal injuries such as contusions, lacerations, and external hematomas, as well as skull fractures. Shock waves from blunt force trauma can also cause underlying focal brain injuries such as cerebral contusions, subarachnoid hematomas, and intracerebral hemorrhages. Sudden movement of the brain relative to the skull induces impulse or inertial loading, which can cause concussion. Inertial loading at the surface of the brain can cause subdural hemorrhage due to bridging vein rupture, whereas its effects on the neural structures deeper within the brain can produce diffuse axonal injury.
Research conducted by Holbourn12 first cited angular (rotational) acceleration as the principal mechanism in brain injury. In studies involving live primates and physical models,8–11,29 Gennarelli, Thibault, and colleagues investigated the importance of rotational acceleration in brain injury causation, concluding that angular acceleration contributes more than linear acceleration to the generation of concussive injuries, diffuse axonal injuries, and subdural hematomas.
Skull Fracture
Thresholds for skull fracture are based on experiments involving 25 gel-filled human cadaveric skulls that were exposed to impacts.22 Each head was filled with gelatin to accurately represent head mass, and the rubber skin of a Hybrid II mannequin covered the skull. A series of 42 frontal, 36 occipital, and 58 temporal blows were delivered to the suspended cadaveric heads, during which linear accelerations were measured. A skull fracture threshold of 250 g was determined for frontal and occipital impacts 3 msec in duration, decreasing to 140 g as the impact duration increased to 7 msec. The skull fracture threshold for lateral impacts was reported as 120 g over 3 msec, decreasing to 90 g as impact duration increased to 7 msec. These findings indicated that skull fracture threshold and impact duration are inversely related.
Concussion
In 1974 Ommaya and Gennarelli investigated the effect of pure translational (linear) or rotational (angular) loading on the heads of primates without inducing contact phenomena.21 It was learned that pure translation produces focal injuries, such as contusions, while rotationally induced inertial loading causes diffuse effects, including concussion and subdural hematoma.
Several studies have been performed to identify thresholds for concussion in football. Pellman and colleagues conducted a series of biomechanical studies in which video-recorded concussive impacts during NFL football games were analyzed.23 The investigators reported that concussions were related to linear and rotational accelerations on the order of 98 ± 28 g and 6432 ± 1813 rad/sec2, respectively. Results showed that a concussive injury is possible at 45 g/3500 rad/sec2, while 5500 rad/sec2 represents a 50% risk of concussive trauma.
Rowson and Duma conducted extensive laboratory and field-based biomechanical evaluations of head injuries in football.24–27 Findings were based on a large database of head impacts recorded using the Head Impact Telemetry System (HITS), developed by Simbex Inc., which utilizes an array of uni-axial accelerometers in the crown of a player’s helmet. Given data for 62,974 subconcussive impacts and 37 diagnosed concussions, the investigators proposed a concussion threshold of 104 ± 30 g and 4726 ± 1931 rad/sec2.26 Their research also offered thresholds for subconcussive impacts, whose cumulative effects, some believe, may be correlated with the early onset of dementia. Average linear and angular accelerations for the upper 50% of subconcussive impacts in the HITS data set consist of impacts of 38 ± 20 g and 1528 ± 984 rad/sec2, respectively, while the top 25% of subconcussive impacts consist of impacts with average accelerations of 52 ± 21 g and 2036 ± 1124 rad/sec2.26
Subdural Hematoma
According to Gennarelli10 the most common type of acute subdural hematoma results from tearing of veins that bridge the subdural space as they travel from the brain’s surface to the various dural sinuses. The severity of injury associated with bridging vein rupture has led to several studies of mechanical failure properties.5,14,15–17,19
Löwenhielm performed mechanical tests on 22 human parasagittal bridging vein samples from 11 persons between the ages of 13 and 87 years with no previous brain injury.15 He hypothesized that when the head is exposed to blunt trauma, the brain is displaced with respect to the dura, causing bridging veins and surrounding connective tissue to be stretched. Löwenhielm stated that maximal shear stresses occur about 7 msec after impact, coinciding with bridging vein disruption. He concluded that bridging vein ruptures occur if the maximal angular acceleration exceeds 4500 rad/sec2.
Depreitere applied a methodology in which 10 unembalmed cadavers were subjected to 18 occipital impacts of varying magnitudes and durations, producing head rotation in the sagittal plane.5 Bridging vein ruptures were detected on autopsy following the injection of contrast dye into the superior sagittal sinus under fluoroscopy. Bridging vein ruptures were produced in 6 impact tests, suggesting a tolerance level of approximately 8000 rad/sec2 for impact durations shorter than 10 msec, which seemed to decrease for longer impact durations. Bridging vein failure data from Depreitere and Löwenhielm are presented in Fig. 1. A line of best fit, which is the curve or mathematical function that has the best fit through all of the data points, was computed, along with the 95% confidence intervals.
Bridging vein failure as a function of peak angular acceleration and impact duration (with a line of best fit and 95% confidence intervals). Figure is available in color online only.
Helmets decrease peak force by extending the duration over which the impact is experienced. In the case of football helmets, the typical impact duration is approximately 15 msec. This suggests that bridging vein rupture may result with peak angular accelerations on the order of 6000 rad/sec2, but may be as low as 4000 rad/sec2 after adjusting for the standard error of the mean in this limited data set (Fig. 1).
Depreitere’s bridging vein rupture threshold is in agreement with epidemiological evidence presented by Forbes et al.,6 who reported an appreciable risk of subdural hematoma associated with head impacts resulting in angular accelerations of 5000 rad/sec2. Furthermore, Forbes and colleagues recently documented a 238% increase in catastrophic head injuries, such as subdural hematomas, particularly among players 18 years old and younger.
In 1985, Ommaya compiled a review of the biomechanics of head injury,20 and his results are included in our summary of biomechanical head and brain injury data in Table 1.
Summary of biomechanical head and brain injury thresholds
| Injury | Source | Description | Linear Acceleration (g) | Angular Acceleration (rad/sec2) |
|---|---|---|---|---|
| Skull fracture | Ono, 1998 | 3 msec—frontal/occipital | 250 | |
| 3 msec—temporal | 120 | |||
| 7 msec—frontal/occipital | 140 | |||
| 7 msec—temporal | 90 | |||
| Subconcussive impact | Ommaya, 1985 | AIS Score 2: moderate | ≤1700 | |
| Pellman et al., 2003 | Average from NFL data set | 57 ± 22 | 4029 ± 1438 | |
| Rowson & Duma, 2013 | Top 50% | 38 ± 20 | 1528 ± 984 | |
| Top 25% | 52 ± 21 | 2036 ± 1124 | ||
| Concussion | Ommaya, 1985 | AIS Score 3: serious | ≤3000 | |
| AIS Score 4: severe | ≤3900 | |||
| Pellman et al., 2003 | Average from NFL data | 98 ± 28 | 6432 ± 1813 | |
| Lowest concussion in NFL | 45 | 3500 | ||
| 50% risk of concussion | 5500 | |||
| Rowson & Duma, 2013 | Average collegiate data | 104 ± 30 | 4726 ± 1931 | |
| Subdural hematoma | Löwenhielm, 1974 | 4500 | ||
| Ommaya, 1985 | AIS Score 5: critical | ≤4500 | ||
| Depreitere et al., 2006 | Calculated for football helmet impact | 4000–6000 | ||
| Forbes et al., 2014 | Epidemiological evidence | 5000 |
It is interesting to note that thresholds for concussive impacts, as calculated by Pellman et al.23 and Rowson and Duma,26 exceed scientific consensus on thresholds for subdural hematoma in football impacts. If such were correct, then the epidemiological data would reflect an epidemic of subdural hematomas, which is not the case. Therefore, one questions the validity of these studies.
Methods
The standard NOCSAE testing system involves a headform, which is rigidly affixed to a metal flyarm. The orientation of the headform can be varied to represent frontal, occipital, lateral, and crown impacts. The flyarm is suspended between 2 parallel wires, which guide the fall of the headform onto an anvil. Since the headform is rigidly affixed to the flyarm without a flexible neck, rotational acceleration is prohibited and thus prevents the objective measurement of concussive risk.
Our modified drop test system incorporates measures of both linear and angular accelerations (Fig. 2). A modification to the NOCSAE standard test apparatus was developed and validated for impact testing of protective head-wear.4 This apparatus incorporates a 50th percentile Hybrid III head and neck assembly from Humanetics ATD. The flexible neck is affixed between the head and flyarm. All other parameters are unchanged. When the headform strikes the anvil, rotational acceleration is unimpeded, thereby permitting measurement of both linear and angular head kinematics in response to the impact.
Modified head drop system with Hybrid III head and neck assembly. Figure is available in color online only.
Instrumentation
A triaxial accelerometer (PCB Piezotronics Inc.) and 3 ARS PRO-18K angular rate sensors (Diversified Technical Systems Inc.) affixed to a triaxial block were installed at the center of mass of the Hybrid III headform (Humanetics ATD). Data from the accelerometer and angular rate sensors were acquired using National Instruments compact DAQ hardware.
Twenty-one football helmet models were evaluated (Table 2). Several models are no longer current, and 5 of these retired models were not available in new condition. Most of the helmets were new, and 3 youth models were included. Except for the MacGregor-Goldsmith leather-head, of which we had only 1 sample, 3 samples of each helmet were evaluated. Each helmet was impacted 5 times in the occipital area from a drop height 2.0 m onto a flat steel anvil, in accordance with ASTM standards.1,2 An impact velocity of 6.2 m/sec (13.9 mph) was computationally verified. Calibration routines were performed at the start and the end of data collection by dropping the Hybrid III headform onto a modular elastomer programmer (MEP) pad 1-in thick with durometer 60 (A scale). An SEM of 0.061 was computed between the pre- and post-calibration tests, indicating high repeatability of the tests.
Football helmets tested
| 1930s MacGregor-Goldsmith Leatherhead*† |
| Adams a2000 Pro*† |
| Rawlings Impulse |
| Rawlings Quantum |
| Rawlings Quantum Plus |
| Riddell Little Pro*†‡ |
| Riddell VSR4*† |
| Riddell Revolution |
| Riddell Revolution Speed |
| Riddell 360 |
| SG Adult Varsity |
| Schutt Air Advantage*† |
| Schutt Ion 4D* |
| Schutt DNA Pro+ |
| Schutt DNA Pro+ youth‡ |
| Schutt Air XP |
| Schutt Air XP Pro |
| Schutt Vengeance DCT |
| Xenith X1 adult* |
| Xenith X1 youth*‡ |
| Xenith X2 adult |
No longer current.
Not available in new condition.
Youth model.
Analysis
In accordance with SAE J211/1,28 data from the analog sensors were acquired at 10,000 Hz per channel using LabVIEW (National Instruments) and then filtered in MATLAB (MathWorks) using a phaseless eighth-order Butterworth filter. Cutoff frequencies of 1650 and 500 Hz were used for the linear accelerometers and angular rate sensors, respectively. Angular acceleration measures were derived from the angular velocity data based on a 5-point least-squares quartic equation. Impact durations were determined based on linear acceleration data.
Results
Table 3 presents a summary of results averaged across 5 tests and 3 samples for each of the 21 helmet models tested. Most helmets met the 140-g skull fracture threshold based on Ono.22 Performance ranged from 94 to 225 g, with an average of 125 g. The Schutt Vengeance ranked first among helmets tested for ability to mitigate linear accelerations. However, the retired Adams a2000 Pro and Schutt Air Advantage helmets, as well as the current Rawlings Quantum and Schutt Air XP helmets, failed to meet this threshold based on an impact velocity of 6.2 m/sec.
In terms of angular (rotational) acceleration, helmet performance ranged from 2628 to 4665 rad/sec2, with an average of 2343 rad/sec2. Of the current varsity helmets, the Rawlings Quantum ranked first in terms of mitigating angular (rotational) accelerations.
Summary of results
| Helmet | Linear Acceleration (g) | GSI | Angular Velocity (rad/sec) | Angular Acceleration (rad/sec2) | Impact Duration (msec) |
|---|---|---|---|---|---|
| None | 1039.1 | 16,639.7 | 14.9 | 9203.5 | 2.1 |
| Adams a2000 Pro*† | 154.1 | 1051.9 | 38.0 | 3754.4 | 13.1 |
| Rawlings Impulse | 111.4 | 637.5 | 33.2 | 3532.4 | 14.5 |
| Rawlings Quantum | 141.1 | 888.3 | 25.0 | 2909.8 | 13.3 |
| Rawlings Quantum Plus | 102.6 | 590.9 | 22.6 | 3290.9 | 15.7 |
| Riddell Little Pro*†‡ | 133.4 | 759.6 | 40.8 | 4348.9 | 14.0 |
| Riddell VSR4*† | 119.0 | 700.4 | 31.6 | 3973.8 | 13.2 |
| Riddell Revolution | 128.3 | 794.1 | 28.4 | 3661.2 | 12.8 |
| Riddell Revolution Speed | 133.8 | 765.6 | 26.8 | 3419.2 | 13.9 |
| Riddell 360 | 131.0 | 726.5 | 23.0 | 3623.1 | 13.2 |
| Schutt Air Advantage*† | 144.9 | 913.7 | 35.6 | 3561.9 | 14.2 |
| Schutt Ion 4D* | 96.0 | 462.0 | 11.1 | 4665.1 | 16.6 |
| Schutt DNA Pro+ | 100.5 | 529.7 | 33.4 | 3113.2 | 20.0 |
| Schutt DNA Pro+ youth‡ | 134.9 | 745.0 | 32.2 | 3651.2 | 11.3 |
| Schutt Air XP | 150.3 | 918.3 | 27.8 | 3297.0 | 12.3 |
| Schutt Air XP Pro | 98.2 | 524.0 | 29.8 | 3131.5 | 21.6 |
| Schutt Vengeance DCT | 93.7 | 486.5 | 25.4 | 3002.2 | 16.8 |
| SG Adult Varsity | 103.9 | 537.2 | 28.2 | 3511.0 | 17.6 |
| Xenith X1 adult* | 105.1 | 599.5 | 34.0 | 3834.4 | 15.4 |
| Xenith X1 youth*‡ | 95.5 | 541.0 | 33.2 | 3788.9 | 18.2 |
| Xenith X2 adult | 119.6 | 710.3 | 34.5 | 4199.9 | 15.6 |
| 1930s MacGregor-Goldsmith leatherhead*† | 225.5 | 1087.6 | 25.9 | 2628.2 | 8.2 |
No longer current.
Not available in new condition.
Youth model.
Figure 3 presents peak angular acceleration versus impact duration for current and retired varsity football helmets. Results for 1 retired football helmet and 3 current varsity helmets—Schutt Ion 4D, Xenith X2, SG Adult Varsity, and Schutt DNA Pro+—fall within the lower confidence level, indicating the possibility of subdural hematoma arising from an impact of 6.2 m/sec. Interestingly, the helmet that seemingly offers the best possible protection against subdural hematoma is the 1930s MacGregor-Goldsmith leatherhead.
Protection against subdural hematoma by current and retired varsity football helmets. Figure is available in color online only.
Given concussion thresholds according to Ommaya,20 Pellman et al.,23 and Rowson and Duma,26 all current varsity football helmets ought to provide adequate protection against the risk of concussion (Fig. 4). However, epidemiological evidence of 300,000+ sports-related concussions in the US annually suggests otherwise.18
Protection against concussion by current and retired football helmets. Figure is available in color online only.
Considering only those varsity helmets currently on the market, we are left with 12 models representing Rawlings, Riddell, Schutt, SG, and Xenith. The Schutt Vengeance offers the best protection against linear accelerations (that is, forces that induce focal head injuries, such as skull fracture), while the Rawlings Quantum offers the best protection against angular (rotational) accelerations (that is, forces that cause concussions and subdural hematomas).
Skull fracture risk was determined for each of the 12 current varsity football helmets by calculating the percent reduction in linear acceleration relative to a 140-g skull fracture threshold, according to Ono.22 Risk of subdural hematoma was determined by calculating the percent reduction in angular acceleration relative to the bridging vein failure threshold computed as a function of impact duration (Fig. 1). Concussion risk was determined by calculating the percent reduction in angular acceleration relative to a 3000-rad/sec2 threshold, according to Ommaya.20
Ranking the helmets according to their performance under these criteria, we determined that the Schutt Vengeance performed the best overall. The ranked summary of results for current varsity football helmets is presented in Fig. 5.
Summary of results for current varsity football helmets (in rank order). Fx = fracture; SDH = subdural hematoma. Figure is available in color online only.
Discussion
Current sports helmet standards fail to address the risks of brain injuries, including concussions and subdural hematomas. The National Operating Committee on Standards for Athletic Equipment (NOCSAE) recently stated, “there is no scientific agreement at all on what the threshold for concussion is.” We believe this confusion is attributable to the inadequacy of several previous studies that have sought to offer concussion thresholds.
In 1985, building upon the research by Löwenhielm,15–17 Ommaya published data on brain injury risk, classified according to the Abbreviated Injury Scale (AIS), as a function of rotational (angular) accelerations.20 He proposed that critical (AIS Score 5) brain injury can be caused by angular accelerations ≥ 4500 rad/sec2, while moderate (AIS Score 2) brain injury can result from angular accelerations > 1700 rad/sec2. Furthermore, Forbes et al. reported that the risk of subdural hematoma becomes appreciable with rotational accelerations of 5000 rad/sec2.6 In contrast, an NFL data set suggested that concussions are associated with angular accelerations of approximately 6432 ± 1813 rad/sec2, and Duma and Rowson suggested a threshold of 4726 ± 1931 rad/sec2 based on their study of concussion in collegiate athletes.23,26
The Pellman23 and Rowson26 studies may be flawed, as their data suggest that limits for concussion are actually higher than those for subdural hematomas, which directly opposes epidemiological evidence.3,18 Further, the validity of the HITS instrumentation, which Rowson and Duma used exclusively in their field testing, has been seriously questioned. Jadischke and colleagues demonstrated that the majority of recordings using this system have an absolute error greater than 15% and that the root mean square error for peak linear acceleration was 59.1%.13 Moreover, the sampling rate of the HITS technology is limited to 1000 Hz, which fails to meet the SAE J21128 guidance for impact measurement. Typical impact duration is on the order of 11–15 msec; therefore, any impact would be grossly under-sampled (that is, there are too few measures). Hence, any proposed thresholds based on this system may by suspect.
We propose helmet protection thresholds to reduce the risk of head and brain injuries in contact sports. Specifically, we recommend that sports helmets be assessed according to the following criteria: 1) To minimize the risk of skull fractures, given research published by Ono,22 we propose a threshold of 140 g for peak linear acceleration to the frontal and occipital areas of the head and 90 g for peak linear acceleration for lateral impacts. 2) To minimize the risk of subdural hematoma, we propose that peak angular accelerations not exceed 50% of the bridging vein failure threshold, calculated as a function of impact duration.5 For example, the average impact duration measured across 12 current varsity football helmets in the present study was 15.2 msec. A bridging vein failure threshold of 4000 rad/sec2 was thus computed based on the equation of the lower confidence interval of the regression line presented in Fig. 1. Allowing for tolerance for intrinsic personal factors that might elevate individual susceptibility, we suggest a peak angular acceleration no greater than 50% of the computed bridging vein failure threshold. 3) To minimize the risk of concussion, we propose a peak angular acceleration threshold of 3000 rad/sec2. 4) To minimize the potential cumulative risk of subconcussive impacts, we propose a peak angular acceleration threshold of 1700 rad/sec2, based on Ommaya’s suggested threshold for moderate brain injury.20 In the interest of simplification and to maximize protection against head and brain injuries to football players and other sports participants of all ages, these thresholds can be summarized as a peak linear acceleration no greater than 90 g and a peak angular acceleration not exceeding 1700 rad/sec2. These proposed thresholds may be refined through future studies.
Pellman23 reported that concussed players experienced head impacts of 9.3 ± 1.9 m/sec (20.8 ± 4.2 mph), which is considerably greater than the impact velocity against which helmets were tested in the present study. We, therefore, propose that helmet-testing standards ought to be revised to reflect impact velocities associated with concussive events.
Conclusions
Our findings demonstrated that not all football helmets provide equal or adequate protection against either focal head injuries or traumatic brain injuries. In fact, some of the most popular helmets on the field ranked among the worst. While protection is improving, none of the current or retired varsity football helmets tested could provide absolute protection against brain injuries, including concussions and subdural hematomas.
It is paramount that protection against traumatic brain injuries be the highest priority. Recent literature suggests that even a single mild traumatic brain injury can have devastating neurological consequences and that players who experience more than 1 mild brain injury are retiring from the game for the sake of their long-term health. To maximize protection against head and brain injuries to football players and other sports participants of all ages, we propose basing thresholds for all sports helmets on a peak linear acceleration no greater than 90 g and a peak angular acceleration not exceeding 1700 rad/sec2.
Author Contributions
Conception and design: both authors. Acquisition of data: Lloyd. Analysis and interpretation of data: both authors. Drafting the article: both authors. Critically revising the article: both authors. Reviewed submitted version of manuscript: both authors. Approved the final version of the manuscript on behalf of both authors: Lloyd.
References
- 1↑
ASTM International: F1446–11a: Standard Test Methods for Equipment and Procedures Used in Evaluating the Performance Characteristics of Protective Headgear West Conshohocken, PA, ASTM International, 2011
- 2↑
ASTM International: F429–10: Standard Test Method for Shock-Attenuation Characteristics of Protective Headgear for Football West Conshohocken, PA, ASTM International, 2007
- 3↑
Boden BP, , Breit I, , Beachler JA, , Williams A, & Mueller FO: Fatalities in high school and college football players. Am J Sports Med 41:1108–1116, 2013
- 4↑
Caccese V, , Ferguson J, , Lloyd J, , Edgecomb M, , Seidi M, & Hajiaghamemar M: Response of an impact test apparatus for fall protective headgear testing using a Hybrid-III head/neck assembly. Exp Tech epub ahead of print 2014
- 5↑
Depreitere B, , Van Lierde C, , Sloten JV, , Van Audekercke R, , Van der Perre G, & Plets C, et al.: Mechanics of acute subdural hematomas resulting from bridging vein rupture. J Neurosurg 104:950–956, 2006
- 6↑
Forbes JA, , Zuckerman S, , Abla AA, , Mocco J, , Bode K, & Eads T: Biomechanics of subdural hemorrhage in American football: review of the literature in response to rise in incidence. Childs Nerv Syst 30:197–203, 2014
- 7↑
Gadd C: Use of a Weighted-Impulse Criterion for Estimating Injury Hazard. SAE Technical Paper 660793 Warrendale, PA, SAE International, 1966
- 8
Gennarelli TA, , Adams JH, & Graham DI: Acceleration induced head injury in the monkey. I The model, its mechanical and physiological correlates Acta Neuropathol Suppl 7:23–25, 1981
- 9
Gennarelli TA, , Ommaya AK, & Thibault LE: Comparison of translational and rotational head motions in experimental cerebral concussion. Stapp Car Crash J 15:797–803, 1971
- 11
Gennarelli TA, , Thibault LE, , Adams JH, , Graham DI, , Thompson CJ, & Marcincin RP: Diffuse axonal injury and traumatic coma in the primate. Ann Neurol 12:564–574, 1982
- 13↑
Jadischke R, , Viano DC, , Dau N, , King AI, & McCarthy J: On the accuracy of the Head Impact Telemetry (HIT) system used in football helmets. J Biomech 46:2310–2315, 2013
- 14↑
Lee MC, & Haut RC: Insensitivity of tensile failure properties of human bridging veins to strain rate: implications in biomechanics of subdural hematoma. J Biomech 22:537–542, 1989
- 16
Löwenhielm P: Strain tolerance of the vv cerebri sup (bridging veins) calculated from head-on collision tests with cadavers. Z Rechtsmed 75:131–144, 1974
- 17
Löwenhielm P: Tolerance level for bridging vein disruption calculated with a mathematical model. J Bioeng 2:501–507, 1978
- 18↑
Marar M, , McIlvain NM, , Fields SK, & Comstock RD: Epidemiology of concussions among United States high school athletes in 20 sports. Am J Sports Med 40:747–755, 2012
- 19↑
Meaney DF: Biomechanics of Acute Subdural Hematoma in the Subhuman Primate and Man [dissertation] Philadelphia, University of Pennsylvania, 1991
- 20↑
Ommaya A, Biomechanics of head injury—experimental aspects. Nahum AM, & Melvin JW: The Biomechanics of Trauma Norwalk, CT, Appleton-Century-Crofts, 1985. 245–269
- 21↑
Ommaya AK, & Gennarelli TA: Cerebral concussion and traumatic unconsciousness. Correlation of experimental and clinical observations of blunt head injuries. Brain 97:633–654, 1974
- 22↑
Ono K, Current status on human head impact tolerance. Yoganandan N, , Pintar FA, , Larson SJ, & Sances A Jr: Frontiers in Head and Neck Trauma: Clinical and Biomechanical Amsterdam, IOS Press, 1998. 183–199
- 23↑
Pellman EJ, , Viano DC, , Tucker AM, , Casson IR, & Waeckerle JF: Concussion in professional football: reconstruction of game impacts and injuries. Neurosurgery 53:799–814, 2003
- 24
Rowson S, , Brolinson G, , Goforth M, , Dietter D, & Duma S: Linear and angular head acceleration measurements in collegiate football. J Biomech Eng 131:061016, 2009
- 25
Rowson S, , Duma SM, , Beckwith JG, , Chu JJ, , Greenwald RM, & Crisco JJ, et al.: Rotational head kinematics in football impacts: an injury risk function for concussion. Ann Biomed Eng 40:1–13, 2012
- 26↑
Rowson S, & Duma SM: Brain injury prediction: assessing the combined probability of concussion using linear and rotational head acceleration. Ann Biomed Eng 41:873–882, 2013
- 27
Rowson S, , Goforth MW, , Dietter D, , Brolinson PG, & Duma SM: Correlating cumulative sub-concussive head impacts in football with player performance - biomed 2009. Biomed Sci Instrum 45:113–118, 2009
- 28↑
ASTM International: J211/1: Instrumentation for Impact Test - Part 1—Electronic Instrumentation Warrendale, PA, SAE International, 2007
- 29↑
Thibault LE, & Gennarelli TA: Biomechanics of diffuse brain injuries Presented at the 29th Stapp Car Crash Conference Washington, DC, 1985. (Abstract) (http://papers.sae.org/856022/) Accessed May 15, 2015
