TY - JOUR
AU - Chancey, Valeta C
AB - Abstract With the prevalence of traumatic brain injury (TBI) in the military and athletics, several commercial and military environmental sensors (ES) have been developed to quantify head impact exposures. The performance of five ES in controlled laboratory exposures from direct and indirect loadings, and the effect on impact protection and dynamic retention of the worn Advanced Combat Helmet (ACH) was evaluated. Direct impacts were conducted on a drop tower and indirect impacts used a mini-sled. The ES data were compared with laboratory sensors through cross-correlation and comparison of peak values. The effects of ES on dynamic retention were assessed using a one-way ANOVA with Tukey’s post hoc analysis against baseline ACH performance. Two ES provided data during the blunt impact tests: one, attached to the side of the headform, correlated well (φ > 0.92) with the laboratory data; the other, mounted in the helmet crown, calculated peak headform velocity, which predicted laboratory velocity well. During indirect impact tests, one environmental sensor (attached to the side of the headform) provided usable data, which correlated well (φ > 0.92) with laboratory data. The inclusion of the environmental sensors did not introduce any safety hazards during the blunt impact attenuation tests or the dynamic retention tests. Concussion, environmental sensor, military training, traumatic brain injury (TBI), mild TBI (mTBI), wearable sensor INTRODUCTION Traumatic brain injury (TBI) has been labeled as the “signature injury” of Operations Iraqi Freedom (OIF) and Enduring Freedom (OEF). It was reported that there were 286,255 mild TBI (mTBI) diagnoses between 2000 and 2016 for all military branches.1 Recently, it has been recognized that at least 80% of the TBI diagnoses were made in a non-deployed (Garrison) setting regardless of where the injury originally occurred.2 TBI diagnoses in the non-deployed setting may be the result of vehicle crashes (private or military-owned), falls, sports and recreational activities, or military training. Early and accurate identification is important for effective concussion management and has implications for extended return-to-duty time lines.3,4 One method for early and accurate identification of a potential concussion is the use of an environmental sensor for measuring head impact events. Research involving the use of impact sensors in sports began nearly 45 yr ago.5,6 More recently, the Head Impact Telemetry System (HITS; Simbex, Lebanon NH, USA) for analyzing impacts in collegiate and high school football has been used to research head impact exposure and concussion injury risk for football players.7–13 The potential use of an environmental sensor for monitoring head impacts has resulted in the development of several new commercial-off-the-shelf (COTS) sensors, including helmet-mounted sensors, mouthguard sensors, headband/skullcap sensors, skin-patch sensors, and inner ear sensors. Few of the new COTS environmental sensors have been used in studies evaluating their performance in the laboratory or in the field.14 Additionally, the studies using COTS environmental sensors do not involve military applications or exposures. Environmental sensors have been used by the military as a tool for recording head impact events resulting from blunt or blast exposures in a deployed environment since 2011. Until recently, the use of environmental sensors in the military has been limited to the second-generation helmet-mounted sensor system (GEN II HMS) and the DARPA Blast Gauge with deployed troops or with Breacher training.15–17 There is interest in using an environmental sensor to monitor additional military training environments for potentially concussive events; however, before COTS and Department of Defense (DoD)-developed sensors can be used in a military setting, they must be evaluated to determine if they will introduce a hazard to the Soldier (e.g., decreased blunt impact protection or increased rotation). The purpose of this study is to evaluate various commercial (military and athletic) and prototype environmental sensors under standard laboratory testing to ensure that they do not introduce an additional safety hazard to the Soldier. Furthermore, the environmental sensor data must be accurate in realistic conditions recreated in the laboratory for the environmental sensor to be useful. Therefore, a secondary objective of this study is to assess the environmental sensor accuracy under direct head exposures (e.g., blunt impact tests) and indirect head exposures (e.g., inertial loading of the head with no direct contact to the head). METHODS All tests were conducted using size large advanced combat helmets (ACHs) per helmet sizing guidelines specified in the instruction manual for the ACH.18 All helmets were configured with new retention systems and Team Wendy® pad sets (installed in the “standard” configuration for use in airborne operations as outlined in the instruction manual for the ACH).18 Currently available military and COTS sensors were assessed for compatibility with military personal protective equipment (PPE), specifically the ACH. Three DoD-developed sensors (A, B, and E) and two COTS sensors (C and D) were identified for use in the present study. Four of the five sensors (A, B, C, and E) were capable of measuring linear accelerations of either the helmet or the head. Three of the five sensors (A, C, and D) were capable of measuring rotational motion of either the helmet or the head. Sensor E (a developmental sensor) was designed to measure contact pressure resulting from impacts. Sensors A, B, D, and E allow data from each event to be downloaded for review (sensor C was limited to LED indicator lights). Only sensors A, B, and D provide full signal time traces that could be used to further analyze the data (Table I). Table I. Overview of Environmental Sensor Characteristics used in the Current Laboratory Evaluation Sensor . Mount Location/ Type . Primary Measurement(s) . Sampling Rate . Data Storage Capacity . Battery Life . Data Output . Sensor A Helmet crown (epoxy) Linear acceleration X, Y, Z 20,000 Hz 1 gigabit (1,000+ events) 12 mo Time trace signals and algorithm that predicts peak headform velocity at the center of gravity with green/amber/red scores Angular rate about the X, Y, Z pressure Sensor B Helmet/body armor (attached to MOLLE straps) Linear acceleration X, Y, Z 20,000 Hz 10 events 6 h to 120 d (in hibernation) LED indicator (green, amber, red) and time trace signals pressure Sensor C Head (skullcap) Linear acceleration X, Y, Z Unknown Not specified 6–7 h LED indicator (green, amber, red) Angular rate about the X, Y, Z Sensor D Head – mastoid process (adhesive) Linear acceleration X, Y, Z 1000 Hz 1,630+ events 6–7 h Time-stamped signal traces and summary statistics (after post- processing) Angular rate about the X, Y, Z Sensor E Helmet (epoxy and adhesive) Contact pressure Unknown 650+ events 12 mo Summary of processed acceleration and pressure metrics from event (peaks, type of event, direction, etc.) Sensor . Mount Location/ Type . Primary Measurement(s) . Sampling Rate . Data Storage Capacity . Battery Life . Data Output . Sensor A Helmet crown (epoxy) Linear acceleration X, Y, Z 20,000 Hz 1 gigabit (1,000+ events) 12 mo Time trace signals and algorithm that predicts peak headform velocity at the center of gravity with green/amber/red scores Angular rate about the X, Y, Z pressure Sensor B Helmet/body armor (attached to MOLLE straps) Linear acceleration X, Y, Z 20,000 Hz 10 events 6 h to 120 d (in hibernation) LED indicator (green, amber, red) and time trace signals pressure Sensor C Head (skullcap) Linear acceleration X, Y, Z Unknown Not specified 6–7 h LED indicator (green, amber, red) Angular rate about the X, Y, Z Sensor D Head – mastoid process (adhesive) Linear acceleration X, Y, Z 1000 Hz 1,630+ events 6–7 h Time-stamped signal traces and summary statistics (after post- processing) Angular rate about the X, Y, Z Sensor E Helmet (epoxy and adhesive) Contact pressure Unknown 650+ events 12 mo Summary of processed acceleration and pressure metrics from event (peaks, type of event, direction, etc.) Open in new tab Table I. Overview of Environmental Sensor Characteristics used in the Current Laboratory Evaluation Sensor . Mount Location/ Type . Primary Measurement(s) . Sampling Rate . Data Storage Capacity . Battery Life . Data Output . Sensor A Helmet crown (epoxy) Linear acceleration X, Y, Z 20,000 Hz 1 gigabit (1,000+ events) 12 mo Time trace signals and algorithm that predicts peak headform velocity at the center of gravity with green/amber/red scores Angular rate about the X, Y, Z pressure Sensor B Helmet/body armor (attached to MOLLE straps) Linear acceleration X, Y, Z 20,000 Hz 10 events 6 h to 120 d (in hibernation) LED indicator (green, amber, red) and time trace signals pressure Sensor C Head (skullcap) Linear acceleration X, Y, Z Unknown Not specified 6–7 h LED indicator (green, amber, red) Angular rate about the X, Y, Z Sensor D Head – mastoid process (adhesive) Linear acceleration X, Y, Z 1000 Hz 1,630+ events 6–7 h Time-stamped signal traces and summary statistics (after post- processing) Angular rate about the X, Y, Z Sensor E Helmet (epoxy and adhesive) Contact pressure Unknown 650+ events 12 mo Summary of processed acceleration and pressure metrics from event (peaks, type of event, direction, etc.) Sensor . Mount Location/ Type . Primary Measurement(s) . Sampling Rate . Data Storage Capacity . Battery Life . Data Output . Sensor A Helmet crown (epoxy) Linear acceleration X, Y, Z 20,000 Hz 1 gigabit (1,000+ events) 12 mo Time trace signals and algorithm that predicts peak headform velocity at the center of gravity with green/amber/red scores Angular rate about the X, Y, Z pressure Sensor B Helmet/body armor (attached to MOLLE straps) Linear acceleration X, Y, Z 20,000 Hz 10 events 6 h to 120 d (in hibernation) LED indicator (green, amber, red) and time trace signals pressure Sensor C Head (skullcap) Linear acceleration X, Y, Z Unknown Not specified 6–7 h LED indicator (green, amber, red) Angular rate about the X, Y, Z Sensor D Head – mastoid process (adhesive) Linear acceleration X, Y, Z 1000 Hz 1,630+ events 6–7 h Time-stamped signal traces and summary statistics (after post- processing) Angular rate about the X, Y, Z Sensor E Helmet (epoxy and adhesive) Contact pressure Unknown 650+ events 12 mo Summary of processed acceleration and pressure metrics from event (peaks, type of event, direction, etc.) Open in new tab Laboratory evaluation testing was used to determine whether the inclusion of environmental sensors would influence the protective performance of the ACH. Blunt impact attenuation tests (direct head exposures) were performed to determine if the addition of environmental sensors degraded the blunt impact energy attenuation performance of the helmet. Dynamic retention tests (indirect head exposures) evaluated the dynamic stability of the helmets, with and without environmental sensors, under an inertial loading. The environmental sensors that provided data (A, B, and D) were also evaluated for accuracy by comparing their output with the output from standard laboratory instrumentation (Table II). Table II. Overview of Laboratory Tests Performed, the Protective Characteristic Being Evaluated and the Supplemental Data Obtained from the Environmental Sensor Under Consideration Laboratory Test . Protective Characteristic . Supplemental Data . Blunt impact attenuation tests Evaluates the performance of the ACH under direct loading using a monorail drop tower and seven impact sites The helmet is dropped at 3 m/s and must attenuate the acceleration seen by the headform to less than 150 G Environmental sensors used during this test will provide data resulting from direct impacts (e.g., a blow to the head during a crash event, during a parachutist landing fall, or other training/operational event) Most environmental sensors should trigger under this type of loading Dynamic retention tests Evaluates the performance of the ACH retention system under inertial loading applied through an impact to the chest The helmet cannot perform worse than a comparable reference helmet when an environmental sensor is added Environmental sensors used during this test will provide data from inertial loads to the head (e.g., falls, crash events with no head contact, parachutist landing falls with no head contact) Most environmental sensors will not trigger under low-energy tests for this condition due to the design of the sensor Laboratory Test . Protective Characteristic . Supplemental Data . Blunt impact attenuation tests Evaluates the performance of the ACH under direct loading using a monorail drop tower and seven impact sites The helmet is dropped at 3 m/s and must attenuate the acceleration seen by the headform to less than 150 G Environmental sensors used during this test will provide data resulting from direct impacts (e.g., a blow to the head during a crash event, during a parachutist landing fall, or other training/operational event) Most environmental sensors should trigger under this type of loading Dynamic retention tests Evaluates the performance of the ACH retention system under inertial loading applied through an impact to the chest The helmet cannot perform worse than a comparable reference helmet when an environmental sensor is added Environmental sensors used during this test will provide data from inertial loads to the head (e.g., falls, crash events with no head contact, parachutist landing falls with no head contact) Most environmental sensors will not trigger under low-energy tests for this condition due to the design of the sensor Open in new tab Table II. Overview of Laboratory Tests Performed, the Protective Characteristic Being Evaluated and the Supplemental Data Obtained from the Environmental Sensor Under Consideration Laboratory Test . Protective Characteristic . Supplemental Data . Blunt impact attenuation tests Evaluates the performance of the ACH under direct loading using a monorail drop tower and seven impact sites The helmet is dropped at 3 m/s and must attenuate the acceleration seen by the headform to less than 150 G Environmental sensors used during this test will provide data resulting from direct impacts (e.g., a blow to the head during a crash event, during a parachutist landing fall, or other training/operational event) Most environmental sensors should trigger under this type of loading Dynamic retention tests Evaluates the performance of the ACH retention system under inertial loading applied through an impact to the chest The helmet cannot perform worse than a comparable reference helmet when an environmental sensor is added Environmental sensors used during this test will provide data from inertial loads to the head (e.g., falls, crash events with no head contact, parachutist landing falls with no head contact) Most environmental sensors will not trigger under low-energy tests for this condition due to the design of the sensor Laboratory Test . Protective Characteristic . Supplemental Data . Blunt impact attenuation tests Evaluates the performance of the ACH under direct loading using a monorail drop tower and seven impact sites The helmet is dropped at 3 m/s and must attenuate the acceleration seen by the headform to less than 150 G Environmental sensors used during this test will provide data resulting from direct impacts (e.g., a blow to the head during a crash event, during a parachutist landing fall, or other training/operational event) Most environmental sensors should trigger under this type of loading Dynamic retention tests Evaluates the performance of the ACH retention system under inertial loading applied through an impact to the chest The helmet cannot perform worse than a comparable reference helmet when an environmental sensor is added Environmental sensors used during this test will provide data from inertial loads to the head (e.g., falls, crash events with no head contact, parachutist landing falls with no head contact) Most environmental sensors will not trigger under low-energy tests for this condition due to the design of the sensor Open in new tab Blunt Impact Attenuation Tests The blunt impact attenuation tests (direct impact head exposures) were performed for the environmental sensors paired with an ACH following a modified version of the Federal Motor Vehicles Safety Standard (FMVSS) 218.19 The modified FMVSS 218 reflects specific needs for testing of military helmets, including impact site and subsequent impacts to helmets (CO/PD-05-04).20 The FMVSS 218 regulations for the test drop tower (Fig. 1), headforms (DOT size “C”), impact surfaces (4.83 cm hemispherical anvil), and data collection standard (specified by the Society of Automotive Engineers [SAE] Standard J211) were followed.21,22 A Denton barrier load cell 4773S1 was used to collect triaxial surface impact forces onto the impact anvil used in all tests. The transmitted headform acceleration was measured with an Endevco (Irvine, CA, USA) 7264B-500 T single-axis accelerometer located at the headform center of mass (CM). Data were recorded at 20,000 Hz on a Diversified Technical Systems (DTS) Inc. (Seal Beach, CA, USA) vehicle docking station with a G5 module. Impact velocity was measured immediately before impact using a flag mounted on the falling mass and a velocimeter. High-speed video was collected at 1,000 frames per second (fps) using a Vision Research (Wayne, NJ, USA) Phantom v9 camera. The acceleration and force data were filtered according to SAE J211, Channel Frequency Class (CFC) 1,000 and CFC 600, respectively.21 The peak headform velocity was calculated from the filtered and integrated headform acceleration. Figure 1. Open in new tabDownload slide Monorail drop tower provides a guided, free fall drop, shown equipped with the hemispherical impact anvil, DOT size “C” headform, and an ACH positioned for a crown impact. The blunt impact test performance requirements for the ACH are based on the peak headform acceleration recorded for a given impact velocity. For all impact locations (crown, front, left, left nape, right, right nape, and rear), the failure criteria for a helmet blunt impact test is a headform peak acceleration greater than 150 G at an impact velocity of 3.0 m/s.21,22 In the blunt impact attenuation test series, each helmet/environmental sensor pair was exposed to a total of 14 impacts; two impacts at seven discrete locations on the shell. The headform orientations tested were in accordance with the testing standard for the ACH (CO/PD-05-04).21 The blunt impact attenuation tests are considered destructive and require that a helmet is removed from use after completing the 14 impacts.20 When available, the signals from the environmental sensors were evaluated against the laboratory sensors by calculating the percent difference in the peaks and calculating the normalized cross-correlation coefficient of common signals. Dynamic Retention Tests The dynamic retention tests (indirect head exposures) were performed on the U.S. Army Aeromedical Research Laboratory (USAARL)-developed mini-sled. The mini-sled is a 3.66-m-long horizontal rail system with a low friction sled (Fig. 2). Mounted to the sled is a Hybrid III biodynamic neck assembly and a Facial and Ocular CountermeasUres for Safety (FOCUS) headform (Humanetics ATD, Plymouth, MI, USA). The FOCUS headform is a biofidelic headform with the anthropometry of a 50th percentile male Soldier. Figure 2. Open in new tabDownload slide USAARL Mini-sled test apparatus allows helmet dynamic retention to be evaluated as configured here with the Hybrid III neck assembly and a FOCUS headform wearing an ACH. To perform a test, the sled is positioned at one end of the rail track and accelerated by the impact of a 45.4 kg pendulum. By varying the pendulum release height and the interfacing impact spring, the acceleration pulse can be tuned for specific tests. Sled motion is arrested approximately 1.3 m from the impact point with a braking system composed of an array of Palmyra brushes. The FOCUS headform was mounted facing forward (toward the pendulum). The dynamic retention tests are considered non-destructive and allow for multiple iterations to be performed on each helmet. The dynamic retention performance of the helmets with and without environmental sensors was assessed at two sled acceleration levels. The “low-energy” condition exposed the sled to a 26 G half-sine pulse with an approximate resultant 4.4 m/s velocity change. The “high-energy” condition exposed the sled to a 39 G half-sine pulse and produced approximately a 5.8 m/s velocity change. The target acceleration levels are based on the Eiband Curves for human tolerance to horizontal acceleration applied to the chest.23 Both energy conditions are within the survivable limits; however, the 39 G half-sine pulse approached the upper limit for human tolerance. High-speed video was collected at 1,000 fps using a Vision Research (Wayne, NJ, USA) Phantom v9 camera. Laboratory data for the dynamic retention tests included triaxial acceleration (Endevco 7264B) and angular rate (DTS ARS 18 K) measured at the FOCUS CM, sled acceleration (Endevco 7264B), and triaxial force measured at the pendulum (Denton Barrier Load Cell). Laboratory data were collected at 50,000 Hz using a Pacific Instruments (Concord, CA, USA) PI580 data acquisition system. The dynamic retention evaluation for the ACH is based on the relative motion of the helmet with respect to the headform. The amount of motion between the helmet and the headform was quantified by measuring the angular displacement of the helmet relative to the angular displacement of the head using high-speed video-tracking. Reflective targets and quad markers were adhered to the helmet and headform to facilitate tracking with TEMA 3.8–004 (Specialised Imaging Inc., Temecula, CA, USA) (Fig. 3A). The rotation of the helmet relative to the headform was calculated using the change in the angle created by the helmet–earcup line and the nose–neck line (Fig. 3B). The amount of motion between the head and helmet was normalized to the initial head–helmet position so that any helmet motion relative to the head would be easily observed. The failure criterion for dynamic retention is based on whether the system under investigation rotates more than a reference system. Figure 3. Open in new tabDownload slide Tracking markers on the FOCUS headform and the ACH (A) were used to create a relative angle between the helmet–earcup line and the nose–neck line (B) for the dynamic retention test video-tracking analysis. A one-way analysis of variance (ANOVA) with 1 × 6 design was performed to provide an additional objective measure of the effects of the environmental sensors on the angular displacement of the helmet. The one-way ANOVA tested the equality of the mean peak-to-peak angular displacement for each environmental sensor/helmet pair and a reference helmet. A post hoc Tukey’s analysis was performed to identify which environmental sensors caused a significantly different peak-to-peak angular displacement compared with the reference (α = 0.05). A statistically significant outcome indicated a difference in the dynamic retention performance of the helmet with an environmental sensor; however, it did not indicate that the helmet/environmental sensor pair failed the safety evaluation. When available, the signals from the environmental sensors were evaluated against the laboratory sensors by calculating the percent difference in the peaks and calculating the normalized cross-correlation coefficient of common signals. Laboratory acceleration data were filtered from the sled (CFC60) and FOCUS CM (CFC1000) before analysis.21 RESULTS Three DoD-developed helmet-mounted sensors (A, B, and E) and two COTS-developed head-mounted sensors (C and D) suitable for use in an acceleration/deceleration environment were assessed to determine whether they introduce a safety hazard to the Soldier when used in the military. The environmental sensors were exposed to direct head exposures (blunt impact attenuation tests) and indirect head exposures (dynamic retention tests). The inclusion of the environmental sensors did not introduce any safety hazards during the blunt impact attenuation tests (no tests exceeded 150 G for the head acceleration) or the dynamic retention tests. Additionally, the environmental sensor performance was evaluated against laboratory data. Several of the sensors did not provide usable data for analysis during either the blunt impact attenuation tests or the dynamic retention tests (sensors B, C, and E). Sensor A had difficulty capturing events resulting from off-axis loading (direct head exposures) or inertial loading (indirect head exposures). Sensor D collected data from both the blunt impact attenuation tests and the dynamic retention tests; however, it often triggered too many times and had a substantial amount of false events. Blunt Impact Attenuation Tests Each blunt impact attenuation test consists of a primary impact against the anvil followed by a rebound and then a smaller, secondary impact (Fig. 4). Three of the five environmental sensors being evaluated triggered during the primary impact of the blunt impact attenuation tests (sensors A, D, and E), and one of the sensors triggered during the secondary impact (sensor D). Sensor A (mounted to the crown of the helmet) reliably triggered and provided data for all impact sites except the left and right nape. The trigger threshold for sensor A was 120 G in each axis, which was not likely to be exceeded when the helmet is impacted at an off-axis location (e.g., the nape locations). As expected, the acceleration recorded by sensor A was not comparable with the laboratory data since it is a helmet-mounted sensor (Fig. 5A); however, the estimated headform velocity (calculated using a proprietary transfer function) corresponded well with the velocity calculated from the laboratory data for all sites where data were available, except the crown (Fig. 5B). Sensor A is located in the crown of the ACH and substantially overpredicts acceleration and headform velocity for impacts directly on the sensor. Figure 4. Open in new tabDownload slide The monorail drop test (A) creates a drop sequence of primary and secondary impacts (B), which one sensor of five was able to capture. Figure 5. Open in new tabDownload slide Sensor A overpredicts accelerations for all sites and overpredicts velocity for the crown impact site. Off-axis sites (left and right nape) did not provide data due to the high trigger threshold. (A) Average peak acceleration from the laboratory headform accelerometer and sensor A. (B) Average peak velocity calculated from the laboratory headform accelerometer and the predicted peak velocity of the head CM provided by the sensor A transfer function in meter per second (m/s). The average peak acceleration recorded by sensor D (Fig. 6) corresponded well with the laboratory headform acceleration. The average percent difference between the headform acceleration and the processed acceleration from sensor D across all impact locations was 3.12%. The normalized cross-correlation coefficient [Φ] was greater than 0.92 for the resultant accelerations (Fig. 7). Sensor E did not provide any usable data for analysis. Figure 6. Open in new tabDownload slide The average peak accelerations recorded by the laboratory and by sensor D were similar for all impact sites. The chart illustrates the variation of results from the laboratory accelerometer data (filtered at CFC1000) and the raw and processed data for sensor D. Figure 7. Open in new tabDownload slide The normalized cross-correlation coefficient [Φ] was greater than 0.92 for the resultant accelerations at all impact sites. Dynamic Retention Tests The dynamic retention tests simulate an impact to the chest (T1-C7 vertebral junction), initially causing the head to rotate forward until maximum neck flexion. The head rotation then changes direction, passing through neutral and continuing a rearward rotation until maximum neck extension. During the described head/neck motion, the helmet can move semi-independently of the head as it is not perfectly coupled (Fig. 8). Figure 8. Open in new tabDownload slide Mini-sled headform motion and event timing: (A) sled and helmet/headform position before impact, at maximum flexion, and at maximum extension; (B) sled acceleration (filtered at CFC60) and FOCUS headform resultant acceleration (filtered at CFC1000); (C) FOCUS headform angular displacement (calculated from FOCUS CM angular rate sensors). (D) rotation of the helmet relative to the headform. The dynamic retention tests using the mini-sled apparatus involve a highly repeatable indirect exposure to the headform. The peak sled acceleration from the low-energy tests was −26.1 ± 0.1 G (n = 9 tests), whereas the peak sled acceleration from the high-energy tests was −38.8 ± 0.1 G (n = 9 tests). The response of the FOCUS headform was also very repeatable. The resultant acceleration at the FOCUS CM for the low-energy tests was −16.4 ± 0.2 G (n = 9) and −27.5 ± 0.5 G (n = 9) for the high-energy tests. The angular rate about the y-axis (pitch) of the headform for the low-energy tests was −1,457.0 ± 32.0 deg/s (n = 9) and −1,9,84.9 ± 35.9 deg/s (n = 9) for the high-energy tests. The repeatability of the mini-sled system allows for comparisons between multiple parameters across all tests at a given energy level. The results from the one-way ANOVA tests for the low-energy and high-energy tests are presented in Tables III and IV. The means for all conditions (reference [no sensor], sensors A, B, C, D, and E) were not equal (p < 0.01 for the low-energy and high-energy tests). For the low-energy tests, the Tukey's post hoc analysis indicated that none of the environmental sensors resulted in a significant difference in the peak-to-peak angular displacement of the helmet when compared with the reference helmet. For the high-energy condition, the tests with sensor B had a significantly reduced peak-to-peak angular displacement of the helmet compared with the reference tests, and the tests with sensor C had a significantly increased peak-to-peak angular displacement of the helmet compared with the reference tests. Table III. One-Way ANOVA Analysis Results for All Low-Energy Mini-sled Tests Helmet . Number of Tests . Average (Degrees) . Standard Deviation . Groupa . Reference 9 26.87 6.44 x y Sensor A 12 26.52 5.40 x y Sensor B 9 21.41 3.15 y Sensor C 9 29.30 2.96 x Sensor D 9 26.47 2.21 x y Sensor E 9 22.92 2.94 y Helmet . Number of Tests . Average (Degrees) . Standard Deviation . Groupa . Reference 9 26.87 6.44 x y Sensor A 12 26.52 5.40 x y Sensor B 9 21.41 3.15 y Sensor C 9 29.30 2.96 x Sensor D 9 26.47 2.21 x y Sensor E 9 22.92 2.94 y aHelmet conditions that do not share a grouping letter are significantly different (i.e., sensors B and C had significantly different ranges of motion, but neither were significantly different from the reference helmet). Open in new tab Table III. One-Way ANOVA Analysis Results for All Low-Energy Mini-sled Tests Helmet . Number of Tests . Average (Degrees) . Standard Deviation . Groupa . Reference 9 26.87 6.44 x y Sensor A 12 26.52 5.40 x y Sensor B 9 21.41 3.15 y Sensor C 9 29.30 2.96 x Sensor D 9 26.47 2.21 x y Sensor E 9 22.92 2.94 y Helmet . Number of Tests . Average (Degrees) . Standard Deviation . Groupa . Reference 9 26.87 6.44 x y Sensor A 12 26.52 5.40 x y Sensor B 9 21.41 3.15 y Sensor C 9 29.30 2.96 x Sensor D 9 26.47 2.21 x y Sensor E 9 22.92 2.94 y aHelmet conditions that do not share a grouping letter are significantly different (i.e., sensors B and C had significantly different ranges of motion, but neither were significantly different from the reference helmet). Open in new tab Table IV. One-Way ANOVA Analysis Results for All High-Energy Tests Helmet . Number of Tests . Average (Degrees) . Standard Deviation . Groupa . Reference 9 34.03 1.32 x y Sensor A 12 35.82 3.25 x y Sensor B 9 29.32 3.04 z Sensor C 9 39.85 2.02 w Sensor D 9 37.42 2.37 w x Sensor E 9 32.92 3.74 y z Helmet . Number of Tests . Average (Degrees) . Standard Deviation . Groupa . Reference 9 34.03 1.32 x y Sensor A 12 35.82 3.25 x y Sensor B 9 29.32 3.04 z Sensor C 9 39.85 2.02 w Sensor D 9 37.42 2.37 w x Sensor E 9 32.92 3.74 y z aHelmet conditions that do not share a grouping letter are significantly different (i.e., sensors B and C were significantly different from the reference tests). Open in new tab Table IV. One-Way ANOVA Analysis Results for All High-Energy Tests Helmet . Number of Tests . Average (Degrees) . Standard Deviation . Groupa . Reference 9 34.03 1.32 x y Sensor A 12 35.82 3.25 x y Sensor B 9 29.32 3.04 z Sensor C 9 39.85 2.02 w Sensor D 9 37.42 2.37 w x Sensor E 9 32.92 3.74 y z Helmet . Number of Tests . Average (Degrees) . Standard Deviation . Groupa . Reference 9 34.03 1.32 x y Sensor A 12 35.82 3.25 x y Sensor B 9 29.32 3.04 z Sensor C 9 39.85 2.02 w Sensor D 9 37.42 2.37 w x Sensor E 9 32.92 3.74 y z aHelmet conditions that do not share a grouping letter are significantly different (i.e., sensors B and C were significantly different from the reference tests). Open in new tab The only sensor that triggered and provided usable data for the mini-sled tests was sensor D. The laboratory resultant acceleration was compared against both the raw and the processed resultant acceleration peaks provided by the sensor (Fig. 9). Additionally, the processed peak angular rate from sensor D was compared against the laboratory angular rate (Fig. 10). The normalized cross-correlation coefficient [Φ] was greater than 0.95 for the resultant acceleration and greater than 0.82 for the angular rate about the y-axis (pitch) for both the low- and the high-energy tests (Fig. 11). Figure 9. Open in new tabDownload slide Average peak resultant accelerations for sensor D (raw and processed) and the laboratory CM accelerometer (filtered at CFC1000) were similar for both energy levels; however, sensor D consistently overpredicted the resultant acceleration. Figure 10. Open in new tabDownload slide Average peak angular rate about the y-axis (pitch) for sensor D and the laboratory CM angular rate sensor about the y-axis were similar for both energy levels. Figure 11. Open in new tabDownload slide Mini-sled normalized average cross-correlation scores between the laboratory grade data and the environmental sensor data were greater than 0.95 for the raw resultant acceleration and greater than 0.82 for the angular rate (pitch). A similar analysis using the processed data was not possible because the processing algorithm only provides peak values. DISCUSSION Three of the five environmental sensors evaluated failed to provide usable data for any analysis beyond assessing potential safety hazards. Sensors A and D provided data that could be used to assess the environmental sensor accuracy; however, they still encountered errors. Sensor A failed to provide data when exposed to impacts not in line with the sensor’s coordinate system (direct head exposures) or inertial loading (indirect head exposures). Conversely, sensor D provided additional data for each test due to a low trigger threshold, which resulted in several false events. Sensors B, C, and E failed to provide usable data for both the direct and indirect head exposure tests. Sensor B was designed to collect primarily pressure data and did not trigger off of acceleration. Sensor C supplied only a visual indicator for severity (green, amber, or red depending on the severity), and it did not trigger reliably for any of the tests (the trigger threshold incorporated both rotational and linear components). Sensor E was a prototype sensor that was unable to provide usable data for any test condition. The dynamic retention assessment of a helmet is a comparative analysis of the rotation of a helmet relative to the headform, usually between a reference (standard) helmet or set of helmets and a modified (environmental sensor-equipped) helmet or set of helmets. Statistical analysis was performed on the peak-to-peak angular displacement (positive peak angle to negative peak angle) for all the helmets tested. The results presented above highlight how the inclusion of an environmental sensor affects the retention, but do not provide an indication of “good versus bad.” The imperfect coupling of the helmet with the head highlighted during the dynamic retention tests may lead to errors in inferences drawn from helmet-based sensors (sensors A, B, and E). One major limitation of all of the environmental sensors used in the present study is the proprietary nature of the processing algorithms used to provide a head exposure. Often, the proprietary algorithms provide a single value (either a peak value or an LED indicator light) that is intended to represent the exposure to the head. Unfortunately, the metrics for relating head exposure to concussion are still being investigated and a single value from an environmental sensor limits future incorporation into a dose–response analysis. Additionally, as the output from the environmental sensors is simplified further, it becomes more difficult to compare data collected in the field (theater or training) back to any data collected in a laboratory that would allow for potential categorization of exposure type (blast, blunt impact, ballistic impact, etc.). The present study has several limitations. The primary goal of the study was to evaluate the effect of environmental sensors on the protective performance of the ACH. This goal allowed some assessment of the accuracy of the environmental sensors as well; however, the tests performed and the available sensors resulted in only two of the five sensors being evaluated providing usable data. The tests performed are unique to the military and may not provide the best representation of how accurate a sensor is under all possible loading conditions. Several studies have evaluated the performance of COTS sensors against standard athletic tests or athletic exposures; however, many of them incorporated a football helmet in their testing, which may change the response compared with an ACH.14,24–29 A major limitation with the use of any environmental sensor is understanding the accuracy of the measurements made using the device. Additional work characterizing the errors associated with the environmental sensor in the laboratory as well as in human subjects is needed. The tests performed for the present study included simple direct blunt and indirect head exposures conducted in a single direction. Accelerative loadings can occur in any axis and may result from direct impacts, indirect impacts, or other complex loading scenarios. Further understanding of the environmental sensor performance under complex loading and off-axis loading is needed. Additionally, accelerative loadings may result from multiple types of threats (e.g., blunt impacts, blast exposures, falls, and vehicle crashes). Finally, the present study only evaluated a subset of environmental sensors. As concussion management becomes more important in sports and the military, environmental sensors for head impact monitoring are being developed and improvements are being made. CONCLUSION The present study evaluated five environmental sensors against standard protection requirements that the ACH is required to meet (blunt impact protection and dynamic retention). The data associated with these tests provided a useful dataset to begin comparing the accuracy and ability of the environmental sensors to collect real events. Although safe for use, three of the five sensors (sensors B, C, and E) were dropped from further consideration as a result of the lack of usable data provided under all test conditions. In addition to the standard protection tests conducted, future work should include evaluations of the environmental sensor’s capabilities in other laboratory-controlled exposure conditions, including complex loading scenarios, projectile impacts, dropped helmet/sensors, and potentially ballistic impacts. Future work should also focus on evaluating the growing population of environmental sensors for head impact monitoring. Presentations Presented as a Poster at the 2016 Military Health System Research Symposium (abstract number: MHSRS-16-1290). Funding This study was supported by an ongoing research for the U.S. Army Medical Research and Materiel Command under the Environmental Sensors in Training (ESiT) Program. REFERENCES 1 Defense and Veterans Brain Injury Center . DoD worldwide numbers for TBI. 2016 . Available at http://dvbic.dcoe.mil/files/tbi-numbers/DoD-TBI-Worldwide-Totals_2000-2016_Q1_May-16-2016_v1.0_2016-06-24.pdf; accessed September 19, 2016. 2 Helmick KM , Spells CA, Malik SZ, Davies CA, Marion DW, Hinds SR: Traumatic brain injury in the US military: epidemiology and key clinical and research programs . Brain Imaging Behav 2015 ; 9 ( 3 ): 358 – 66 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Asken BM , McCrea MA, Clugston JR, Snyder AR, Houck ZM, Bauer RM: Playing through it: delayed reporting and removal from athletic activity after concussion predicts prolonged recovery . J Athl Train 2016 ; 51 ( 4 ): 329 – 35 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Meehan WP III , Mannix R, Monuteaux MC, et al. : Early symptom burden predicts recovery after sport-related concussion . Neurology 2014 ; 83 ( 24 ): 2204 – 10 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Moon DW , Beedle CW, Kovacic CR: Peak head acceleration of athletes during competition – football . Med Sci Sports 1971 ; 3 ( 1 ): 44 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 6 Reid SE , Epstein HM, O’Dea TJ, et al. : Head protection in football . Am J Sports Med 1974 ; 2 ( 2 ): 86 – 92 . Google Scholar Crossref Search ADS WorldCat 7 Broglio SP , Sosnoff JJ, Sunhoon S, et al. : Head impacts during high school football: a biomechanical assessment . J Athl Train 2009 ; 44 ( 4 ): 342 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Crisco JJ , Wilcox BJ, Beckwith JG, et al. : Head impact exposure in collegiate football players . J Biomech 2011 ; 44 ( 15 ): 2673 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Duma SM , Manoogian SJ, Bussone WR, et al. : Analysis of real-time head accelerations in collegiate football players . Clin J Sport Med 2005 ; 15 ( 1 ): 3 – 8 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Greenwald RM , Gwin JT, Chu JJ, et al. : Head impact severity measures for evaluating mild traumatic brain injury risk exposure . Neurosurgery 2008 ; 62 ( 4 ): 789 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Guskiewicz KM , Mihalik JP, Shankar V, et al. : Measurement of head impacts in collegiate football players: relationship between head impact biomechanics and acute clinical outcome after concussion . Neurosurgery 2007 ; 61 ( 6 ): 1244 – 53 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Beckwith JG , Greenwald RM, Chu JJ: Measuring head kinematics in football: correlation between the head impact telemetry system and hybrid III headform . Ann Biomed Eng 2012 ; 40 ( 1 ): 237 – 48 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Rowson S , Duma SM, Beckwith JG, et al. : Rotational head kinematics in football impacts: an injury risk function for concussion . Ann Biomed Eng 2012 ; 40 ( 1 ): 1 – 13 . Google Scholar Crossref Search ADS PubMed WorldCat 14 Williams RM , Dowling M, O’Connor KL: Head impact measurement devices: a clinical review. Sports Health: A Multidisciplinary Approach 2016. Available at http://www.safetylit.org/citations/index.php?fuseaction=citations.viewdetails&citationIds[]=citjournalarticle_517735_11; accessed April 26, 2017. 15 Carr W , Yarnell AM, Ong R, et al. : Ubiquitin carboxy-terminal hydrolase l1 as a serum neurotrauma biomarker for exposure to occupational low-level blast . Front Neurol 2015 ; 6 : 1 – 11 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Carr W , Taylor M, LoPresti ML, et al. : Symptomology observed in humans following acute exposure to explosive blast . J Neurotrauma 2015 ; 32 : A-109 . Google Scholar Crossref Search ADS WorldCat 17 LoPresti M , Carr W, Yarnell AM, et al. : An evaluation of the neurological effects of repeated exposure to low-level blast overpressure during U.S. Army explosives training. program No. 703.18. Neuroscience 2014 Abstracts. Washington, DC: Society for Neuroscience, 2014. Available at https://www.sfn.org/Annual-Meeting/Neuroscience-2014/Sessions-and-Events/Program/Abstract-PDFs-2014; accessed 22 May 17. Click on Wednesday poster presentations. 18 Department of the Army : Advanced combat helmet. Washington, D.C. TM 10-8470-204–10, 2004 . Available at http://www.operation-helmet.org/documents/ACH-TM10847020410R001.pdf; accessed April 26, 2017. 19 Department of Transportation : Motorcycle helmets. Federal Motor Vehicle Safety Standard 218, 2006 . Available at https://www.nhtsa.gov/sites/nhtsa.dot.gov/files/fmvss/Helmets_Upgrade_NPRM.pdf; accessed April 26, 2017. 20 Department of the Army : Purchase description helmet, advanced combat (ACH), CO/PD-05-04. 2007 . Fort Belvoir, VA: Project Manager - Soldier Equipment, Executive Office - older. Available at http://www.gentexcorp.com/assets/base/TechnicalPublications/CO-PD-05-041.pdf; accessed April 26, 2017. 21 Society of Automotive Engineers : Instrumentation for impact test-part 1-electronic instrumentation. 1995 . Standard 1995; J211. Available at http://standards.sae.org/j211/1_199503/; accessed April 26, 2017. 22 Vasquez KB , Logsdon KP, Dorman DB, et al. : Combat helmet-headform coupling characterized from blunt impact events. Proceedings of ASME 2011 IMECE. November 11–17 2011 . Denver, CO. Available at http://proceedings.asmedigitalcollection.asme.org/proceeding.aspx?articleid=1642319; accessed April 26, 2017. 23 Eiband AM : Human tolerance to rapidly applied accelerations: a summary of the literature. 1959 . Available at https://archive.org/details/nasa_techdoc_19980228043; accessed April 26, 2017. 24 Siegmund GP , Guskiewicz KM, Marshall SW, DeMarco AL, Bonin SJ: Laboratory validation of two wearable sensor systems for measuring head impact severity in football players . Ann Biomed Eng 2016 ; 44 ( 4 ): 1257 – 74 . Google Scholar Crossref Search ADS PubMed WorldCat 25 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 2013 ; 46 ( 13 ): 2310 – 5 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Wu LC , Laksari K, Kuo C, et al. : Bandwidth and sample rate requirements for wearable head impact sensors . J Biomech 2016 ; 49 ( 13 ): 2918 – 24 . Google Scholar Crossref Search ADS PubMed WorldCat 27 Camarillo DB , Shull PB, Mattson J, Shultz R, Garza D: An instrumented mouthguard for measuring linear and angular head impact kinematics in American football . Ann Biomed Eng 2013 ; 41 ( 9 ): 1939 – 49 . Google Scholar Crossref Search ADS PubMed WorldCat 28 Bartsch A , Samorezov S, Benzel E, Miele V, Brett D: Validation of an “intelligent mouthguard” single event head impact dosimeter . Stapp Car Crash J 2014 ; 58 : 1 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 29 Wu LC , Nangia V, Bui K, et al. : In vivo evaluation of wearable head impact sensors . Ann Biomed Eng 2016 ; 44 ( 4 ): 1234 – 45 . Google Scholar Crossref Search ADS PubMed WorldCat Published by Oxford University Press on behalf of the Association of Military Surgeons of the United States 2018. This work is written by (a) US Government employee(s) and is in the public domain in the US. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) Published by Oxford University Press on behalf of the Association of Military Surgeons of the United States 2018. This work is written by (a) US Government employee(s) and is in the public domain in the US.
TI - Evaluation of Environmental Sensors During Laboratory Direct and Indirect Head Exposures
JF - Military Medicine
DO - 10.1093/milmed/usx208
DA - 2018-03-01
UR - https://www.deepdyve.com/lp/oxford-university-press/evaluation-of-environmental-sensors-during-laboratory-direct-and-2YFZ4lLWsI
SP - 294
EP - 302
VL - 183
IS - suppl_1
DP - DeepDyve
ER -