AI-based denoising of head impact kinematics measurements with convolutional neural network for traumatic brain injury prediction.

OBJECTIVE: Wearable devices are developed to measure head impact kinematics but are intrinsically noisy because of the imperfect interface with human bodies. This study aimed to improve the head impact kinematics measurements obtained from instrumented mouthguards using deep learning to enhance traumatic brain injury (TBI) risk monitoring. METHODS: We developed one-dimensional convolutional neural network (1D-CNN) models to denoise mouthguard kinematics measurements for tri-axial linear acceleration and tri-axial angular velocity from 163 laboratory dummy head impacts. The performance of the denoising models was evaluated on three levels: kinematics, brain injury criteria, and tissue-level strain and strain rate. Additionally, we performed a blind test on an on-field dataset of 118 college football impacts and a test on 413 post-mortem human subject (PMHS) impacts. RESULTS: On the dummy head impacts, the denoised kinematics showed better correlation with reference kinematics, with relative reductions of 36% for pointwise root mean squared error and 56% for peak absolute error. Absolute errors in six brain injury criteria were reduced by a mean of 82%. For maximum principal strain and maximum principal strain rate, the mean error reduction was 35% and 69%, respectively. On the PMHS impacts, similar denoising effects were observed and the peak kinematics after denoising were more accurate (relative error reduction for 10% noisiest impacts was 75.6%). CONCLUSION: The 1D-CNN denoising models effectively reduced errors in mouthguard-derived kinematics measurements on dummy and PMHS impacts. SIGNIFICANCE: This study provides a novel approach for denoising head kinematics measurements in dummy and PMHS impacts, which can be further validated on more real-human kinematics data before real-world applications.

Capturing Head Impacts in Boxing: A Video-Based Comparison of Three Wearable Sensors.

Wearable sensors are used to quantify head impacts in athletes, but recent work has shown that the number of events recorded may not be accurate. This study aimed to compare the number of head acceleration events recorded by three wearable sensors during boxing and assess how impact type and location affect the triggering of acceleration events. Seven boxers were equipped with an instrumented mouthguard, a skin patch, and a headgear patch. Contacts to participants' heads were identified via three video cameras over 115 sparring rounds. The resulting 5168 video-identified events were used as reference to quantify the sensitivity, specificity, and positive predictive value (PPV) of the sensors. The mouthguard, skin patch, and headgear patch recorded 695, 1579, and 1690 events, respectively, yielding sensitivities of 35%, 86%, and 78%, respectively, and specificities of 90%, 76%, and 75%, respectively. The mouthguard, skin patch, and headgear patch yielded 693, 1571, and 1681 true-positive events, respectively, leading to PPVs for head impacts over 96%. All three sensors were more likely to be triggered by punches landing near the sensor and cleanly on the head, although the mouthguard's sensitivity to impact location varied less than the patches. While the use of head impact sensors for assessing injury risks remains uncertain, this study provides valuable insights into the capabilities and limitations of these sensors in capturing video-verified head impact events.

Toward more accurate and generalizable brain deformation estimators for traumatic brain injury detection with unsupervised domain adaptation.

Machine learning head models (MLHMs) are developed to estimate brain deformation for early detection of traumatic brain injury (TBI). However, the overfitting to simulated impacts and the lack of generalizability caused by distributional shift of different head impact datasets hinders the broad clinical applications of current MLHMs. We propose brain deformation estimators that integrates unsupervised domain adaptation with a deep neural network to predict whole-brain maximum principal strain (MPS) and MPS rate (MPSR). With 12,780 simulated head impacts, we performed unsupervised domain adaptation on on-field head impacts from 302 college football (CF) impacts and 457 mixed martial arts (MMA) impacts using domain regularized component analysis (DRCA) and cycle-GAN-based methods. The new model improved the MPS/MPSR estimation accuracy, with the DRCA method significantly outperforming other domain adaptation methods in prediction accuracy (p<0.001): MPS RMSE: 0.027 (CF) and 0.037 (MMA); MPSR RMSE: 7.159 (CF) and 13.022 (MMA). On another two hold-out testsets with 195 college football impacts and 260 boxing impacts, the DRCA model significantly outperformed the baseline model without domain adaptation in MPS and MPSR estimation accuracy (p<0.001). The DRCA domain adaptation reduces the MPS/MPSR estimation error to be well below TBI thresholds, enabling accurate brain deformation estimation to detect TBI in future clinical applications.

Padded Helmet Shell Covers in American Football: A Comprehensive Laboratory Evaluation with Preliminary On-Field Findings.

Protective headgear effects measured in the laboratory may not always translate to the field. In this study, we evaluated the impact attenuation capabilities of a commercially available padded helmet shell cover in the laboratory and on the field. In the laboratory, we evaluated the padded helmet shell cover's efficacy in attenuating impact magnitude across six impact locations and three impact velocities when equipped to three different helmet models. In a preliminary on-field investigation, we used instrumented mouthguards to monitor head impact magnitude in collegiate linebackers during practice sessions while not wearing the padded helmet shell covers (i.e., bare helmets) for one season and whilst wearing the padded helmet shell covers for another season. The addition of the padded helmet shell cover was effective in attenuating the magnitude of angular head accelerations and two brain injury risk metrics (DAMAGE, HARM) across most laboratory impact conditions, but did not significantly attenuate linear head accelerations for all helmets. Overall, HARM values were reduced in laboratory impact tests by an average of 25% at 3.5 m/s (range: 9.7 to 39.6%), 18% at 5.5 m/s (range: - 5.5 to 40.5%), and 10% at 7.4 m/s (range: - 6.0 to 31.0%). However, on the field, no significant differences in any measure of head impact magnitude were observed between the bare helmet impacts and padded helmet impacts. Further laboratory tests were conducted to evaluate the ability of the padded helmet shell cover to maintain its performance after exposure to repeated, successive impacts and across a range of temperatures. This research provides a detailed assessment of padded helmet shell covers and supports the continuation of in vivo helmet research to validate laboratory testing results.

Head Impact Research Using Inertial Sensors in Sport: A Systematic Review of Methods, Demographics, and Factors Contributing to Exposure.

BACKGROUND: The number and magnitude of head impacts have been assessed in-vivo using inertial sensors to characterise the exposure in various sports and to help understand their potential relationship to concussion. OBJECTIVES: We aimed to provide a comprehensive review of the field of in-vivo sensor acceleration event research in sports via the summary of data collection and processing methods, population demographics and factors contributing to an athlete's exposure to sensor acceleration events. METHODS: The systematic search resulted in 185 cohort or cross-sectional studies that recorded sensor acceleration events in-vivo during sport participation. RESULTS: Approximately 5800 participants were studied in 20 sports using 18 devices that included instrumented helmets, headbands, skin patches, mouthguards and earplugs. Female and youth participants were under-represented and ambiguous results were reported for these populations. The number and magnitude of sensor acceleration events were affected by a variety of contributing factors, suggesting sport-specific analyses are needed. For collision sports, being male, being older, and playing in a game (as opposed to a practice), all contributed to being exposed to more sensor acceleration events. DISCUSSION: Several issues were identified across the various sensor technologies, and efforts should focus on harmonising research methods and improving the accuracy of kinematic measurements and impact classification. While the research is more mature for high-school and collegiate male American football players, it is still in its early stages in many other sports and for female and youth populations. The information reported in the summarised work has improved our understanding of the exposure to sport-related head impacts and has enabled the development of prevention strategies, such as rule changes. CONCLUSIONS: Head impact research can help improve our understanding of the acute and chronic effects of head impacts on neurological impairments and brain injury. The field is still growing in many sports, but technological improvements and standardisation of processes are needed.

An Accessible, 16-Week Neck Strength Training Program Improves Head Kinematics Following Chest Perturbation in Young Soccer Athletes.

CONTEXT: Neck size and strength may be associated with head kinematics and concussion risks. However, there is a paucity of research examining neck strengthening and head kinematics in youths. In addition, neck training is likely lacking in youth sport due to a perceived inadequacy of equipment or time. OBJECTIVE: Examine neck training effects with minimal equipment on neck strength and head kinematics following chest perturbations in youth athletes. DESIGN: Single-group, pretest-posttest case series. SETTING: Athlete training center. PARTICIPANTS: Twenty-five (14 men and 11 women) youth soccer athletes (9.8 [1.5] y). INTERVENTION: Sixteen weeks of twice-weekly neck-focused resistance training utilizing bands, body weight, and manual resistance. MAIN OUTCOME MEASURES: Head kinematics (angular range of motion, peak anterior-posterior linear acceleration, and peak resultant linear acceleration) were measured by an inertial motion unit fixed to the apex of the head during torso perturbations. Neck-flexion and extension strength were assessed using weights placed on the forehead and a plate-loaded neck harness, respectively. Neck length and circumference were measured via measuring tape. RESULTS: Neck extension (increase in median values for all: +4.5 kg, +100%, P < .001; females: +4.5 kg, +100%, P = .002; males: +2.2 kg, +36%, P = .003) and flexion (all: +3.6 kg, +114%, P < .001; females: +3.6 kg, +114%, P = .004; males: +3.6 kg, +114%, P = .001) strength increased following the intervention. Men and women both experienced reduced perturbation-induced head pitch (all: -84%, P < .001). However, peak resultant linear acceleration decreased in the female (-53%, P = .004), but not male (-31%, P = 1.0) subgroup. Preintervention peak resultant linear acceleration and extension strength (R2 = .21, P = .033) were the closest-to-significance associations between head kinematics and strength. CONCLUSIONS: Young athletes can improve neck strength and reduce perturbation-induced head kinematics following a 16-week neck strengthening program. However, further research is needed to determine the effect of improved strength and head stabilization on concussion injury rates.

Assessing Head/Neck Dynamic Response to Head Perturbation: A Systematic Review.

BACKGROUND: Head/neck dynamic response to perturbation has been proposed as a risk factor for sports-related concussion. OBJECTIVES: The aim of this systematic review was to compare methodologies utilised to assess head/neck dynamic response to perturbation, report on magnitude, validity and reliability of the response, and to describe modifying factors. METHODS: A systematic search of databases resulted in 19 articles that met the inclusion and exclusion criteria. RESULTS: Perturbation methods for head/neck dynamic response included load dropping, quick release and direct impact. Magnitudes of perturbation energy varied from 0.1 to 11.8 J. Head/neck response was reported as neck muscle latency (18.6-88.0 ms), neck stiffness (147.2-721.9 N/rad, 14-1145.3 Nm/rad) and head acceleration (0.2-3.8g). Reliability was only reported in two studies. Modifying factors for head/neck response included younger and older participants presenting increased responses, females showing better muscular reactivity but similar or increased head kinematics compared with males, and bracing for impact limiting muscular activity and head kinematics. DISCUSSION: Substantial differences in experimental and reporting methodologies limited comparison of results. Methodological factors such as impact magnitude should be considered in future research. CONCLUSION: Each methodology provides valuable information but their validity for anticipated and unanticipated head impacts measured in vivo needs to be addressed. Reports on head/neck response should include measurement of transmitted force, neck muscle latency, head linear and rotational accelerations, and neck stiffness. Modifying factors of anticipation, participants' age, sex, and sport are to be considered for head/neck dynamic response. PROSPERO REGISTRATION NUMBER: CRD42016051057 (last updated on 27 February 2017).