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

Flao, Enora Le; Siegmund, Gunter P.; Borotkanics, Robert

doi:10.1007/s40279-021-01574-y

Cited by 16 publications

(15 citation statements)

References 256 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A potential explanation for the superior performance of our learning system might lie in the fact that both Motiwale et al 29 and DiCesare et al 30 considered other types of impacts, such as ground contacts or body collisions, triggering the sensor and that the kinematic waveforms of these events might substantially differ from the characteristic acceleration profiles of direct head impacts from heading the ball. Taken together, our findings are in line with previous studies by further demonstrating the ability of intelligent, data-driven approaches in differentiating between true headers and non-header events – especially when compared to the employment of sensor manufacturers’ proprietary classification algorithms 15 . However, to exploit the general potential of these models for an automatized detection and classification of soccer headers, future research should focus on the development of learning systems specifically able to handle the considerable class imbalance, which is typically to be observed in sensor data collected in on-field scenarios.…”

Section: Discussionsupporting

confidence: 91%

“…Next to video-based or direct observation of play and self-report techniques, wearable head impact sensors constitute a popular means for the assessment of soccer players’ individual heading exposure 15 – 17 . In the present study, the xPatch sensors captured 904 of 1016 video-identified header events (89% sensitivity) at a pre-determined linear acceleration threshold of 8 g, indicating their potential for an automatized detection of soccer headers in on-field scenarios.…”

Section: Discussionmentioning

confidence: 99%

“…Next to video-based observation of play, which is considered the gold standard 13 , 14 , recent technological advances have contributed to the development of wearable sensor technologies that enable the study of head impacts, along with their associated kinematics, in vivo. As these sensors allow for an automatized registration and characterization of impact events, they have proven useful for studying long-term exposure to RHI and, thus, constitute a popular means for the assessment of soccer players’ individual heading frequency 15 – 17 . However, various studies have shown that available sensor systems suffer from poor accuracy in differentiating true head impacts from other acceleration events, such as jumping or manual contacts with the sensor, that exceed the sensors’ pre-defined trigger threshold (typically 10 g).…”

Section: Introductionmentioning

confidence: 99%

“…In an attempt to minimize the number of false positive recordings, some authors suggested to increase the sensors’ pre-set trigger to 16 g 19 or even 20 g 21 in order to selectively capture meaningful head impacts. However, since many soccer headers result in head accelerations well below these thresholds 18 , 22 , a considerable proportion would be missed by the sensor when employing these recommendations 15 , 17 . Moreover, manufacturers equipped their sensors with proprietary processing algorithms to detect true impacts and remove “spurious” events 19 .…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A neural network for the detection of soccer headers from wearable sensor data

Kern

Lober

Hermsdörfer

et al. 2022

Sci Rep

View full text Add to dashboard Cite

To investigate the proposed association between soccer heading and deleterious brain changes, an accurate quantification of heading exposure is crucial. While wearable sensors constitute a popular means for this task, available systems typically overestimate the number of headers by poorly discriminating true impacts from spurious recordings. This study investigated the utility of a neural network for automatically detecting soccer headers from kinematic time series data obtained by wearable sensors. During 26 matches, 27 female soccer players wore head impacts sensors to register on-field impact events (> 8 g), which were labelled as valid headers (VH) or non-headers (NH) upon video review. Of these ground truth data, subsets of 49% and 21% each were used to train and validate a Long Short-Term Memory (LSTM) neural network in order to classify sensor recordings as either VH or NH based on their characteristic linear acceleration features. When tested on a balanced dataset comprising 271 VHs and NHs (which corresponds to 30% and 1.4% of ground truth VHs and NHs, respectively), the network showed very good overall classification performance by reaching scores of more than 90% across all metrics. When testing was performed on an unbalanced dataset comprising 271 VHs and 5743 NHs (i.e., 30% of ground truth VHs and NHs, respectively), as typically obtained in real-life settings, the model still achieved over 90% sensitivity and specificity, but only 42% precision, which would result in an overestimation of soccer players’ true heading exposure. Although classification performance suffered from the considerable class imbalance between actual headers and non-headers, this study demonstrates the general ability of a data-driven deep learning network to automatically classify soccer headers based on their linear acceleration profiles.

show abstract

Section: Discussionsupporting

confidence: 91%

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A neural network for the detection of soccer headers from wearable sensor data

Kern

Lober

Hermsdörfer

et al. 2022

Sci Rep

View full text Add to dashboard Cite

show abstract

“…94 When choosing data acquisition parameters, consider the duration and minimum severity of the test conditions. Wearable devices typically use a threshold based on linear acceleration to trigger data collection (e.g., 10-15 g); 21 however, use of a trigger based on angular kinematics may also be needed in some conditions, because linear acceleration varies across a rigid body during 6DOF motion. 81,92 When selecting parameters for the recording duration, ensure that sufficient data are collected to accomplish the study objectives.…”

Section: Specific Considerationsmentioning

confidence: 99%

Consensus Head Acceleration Measurement Practices (CHAMP): Laboratory Validation of Wearable Head Kinematic Devices

Gabler¹,

Patton

Begonia

et al. 2022

Ann Biomed Eng

View full text Add to dashboard Cite

Wearable devices are increasingly used to measure real-world head impacts and study brain injury mechanisms. These devices must undergo validation testing to ensure they provide reliable and accurate information for head impact sensing, and controlled laboratory testing should be the first step of validation. Past validation studies have applied varying methodologies, and some devices have been deployed for on-field use without validation. This paper presents best practices recommendations for validating wearable head kinematic devices in the laboratory, with the goal of standardizing validation test methods and data reporting. Key considerations, recommended approaches, and specific considerations were developed for four main aspects of laboratory validation, including surrogate selection, test conditions, data collection, and data analysis. Recommendations were generated by a group with expertise in head kinematic sensing and laboratory validation methods and reviewed by a larger group to achieve consensus on best practices. We recommend that these best practices are followed by manufacturers, users, and reviewers to conduct and/or review laboratory validation of wearable devices, which is a minimum initial step prior to on-field validation and deployment. We anticipate that the best practices recommendations will lead to more rigorous validation of wearable head kinematic devices and higher accuracy in head impact data, which can subsequently advance brain injury research and management.

show abstract

Embracing the impact from instrumented mouthguards (iMGs): A survey of iMG managers' perceptions of staff and player interest into the technology, data and barriers to use

Roe,

Whitehead,

Starling

et al. 2024

European Journal of Sport Science

View full text Add to dashboard Cite

Instrumented mouthguards (iMGs) are a novel technology being used within rugby to quantify head acceleration events. Understanding practitioners' perceptions of the barriers and facilitators to their use is important to support implementation and adoption. This study assessed men's and women's rugby union and league iMG managers' perceptions of staff and player interest in the technology, data and barriers to use. Forty‐six iMG managers (men's rugby union and league n = 20 and n = 9 and women's rugby union and league n = 7 and n = 10) completed an 18‐question survey. Perceived interest in data varied across staff roles with medical staff being reported as having the most interest. The iMG devices were perceived as easy to use but uncomfortable. Several uses of data were identified, including medical applications, player monitoring and player welfare. The comfort, size and fit of the iMG were reported as the major barriers to player use. Time constraints and a lack of understanding of data were barriers to engagement with the data. Continued education on how iMG data can be used is required to increase player and staff buy‐in, alongside improving comfort of the devices. Studies undertaken with iMGs investigating player performance and welfare outcomes will make data more useful and increase engagement.

show abstract

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

Cited by 16 publications

References 256 publications

A neural network for the detection of soccer headers from wearable sensor data

A neural network for the detection of soccer headers from wearable sensor data

Consensus Head Acceleration Measurement Practices (CHAMP): Laboratory Validation of Wearable Head Kinematic Devices

Embracing the impact from instrumented mouthguards (iMGs): A survey of iMG managers' perceptions of staff and player interest into the technology, data and barriers to use

Contact Info

Product

Resources

About