Sensors

Research

Jump to: Other

19 pages, 15402 KiB

Open AccessArticle

Drift-Free 3D Orientation and Displacement Estimation for Quasi-Cyclical Movements Using One Inertial Measurement Unit: Application to Running

by Marit A. Zandbergen, Jasper Reenalda, Robbert P. van Middelaar, Romano I. Ferla, Jaap H. Buurke and Peter H. Veltink

Sensors 2022, 22(3), 956; https://doi.org/10.3390/s22030956 - 26 Jan 2022

Cited by 7 | Viewed by 3394

Abstract

A Drift-Free 3D Orientation and Displacement estimation method (DFOD) based on a single inertial measurement unit (IMU) is proposed and validated. Typically, body segment orientation and displacement methods rely on a constant- or zero-velocity point to correct for drift. Therefore, they are not [...] Read more.

A Drift-Free 3D Orientation and Displacement estimation method (DFOD) based on a single inertial measurement unit (IMU) is proposed and validated. Typically, body segment orientation and displacement methods rely on a constant- or zero-velocity point to correct for drift. Therefore, they are not easily applicable to more proximal segments than the foot. DFOD uses an alternative single sensor drift reduction strategy based on the quasi-cyclical nature of many human movements. DFOD assumes that the quasi-cyclical movement occurs in a quasi-2D plane and with an approximately constant cycle average velocity. DFOD is independent of a constant- or zero-velocity point, a biomechanical model, Kalman filtering or a magnetometer. DFOD reduces orientation drift by assuming a cyclical movement, and by defining a functional coordinate system with two functional axes. These axes are based on the mean acceleration and rotation axes over multiple complete gait cycles. Using this drift-free orientation estimate, the displacement of the sensor is computed by again assuming a cyclical movement. Drift in displacement is reduced by subtracting the mean value over five gait cycle from the free acceleration, velocity, and displacement. Estimated 3D sensor orientation and displacement for an IMU on the lower leg were validated with an optical motion capture system (OMCS) in four runners during constant velocity treadmill running. Root mean square errors for sensor orientation differences between DFOD and OMCS were 3.1 ± 0.4° (sagittal plane), 5.3 ± 1.1° (frontal plane), and 5.0 ± 2.1° (transversal plane). Sensor displacement differences had a root mean square error of 1.6 ± 0.2 cm (forward axis), 1.7 ± 0.6 cm (mediolateral axis), and 1.6 ± 0.2 cm (vertical axis). Hence, DFOD is a promising 3D drift-free orientation and displacement estimation method based on a single IMU in quasi-cyclical movements with many advantages over current methods. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

26 pages, 6043 KiB

Open AccessArticle

Fusion Models for Generalized Classification of Multi-Axial Human Movement: Validation in Sport Performance

by Rajesh Amerineni, Lalit Gupta, Nathan Steadman, Keshwyn Annauth, Charles Burr, Samuel Wilson, Payam Barnaghi and Ravi Vaidyanathan

Sensors 2021, 21(24), 8409; https://doi.org/10.3390/s21248409 - 16 Dec 2021

Cited by 8 | Viewed by 2693

Abstract

We introduce a set of input models for fusing information from ensembles of wearable sensors supporting human performance and telemedicine. Veracity is demonstrated in action classification related to sport, specifically strikes in boxing and taekwondo. Four input models, formulated to be compatible with [...] Read more.

We introduce a set of input models for fusing information from ensembles of wearable sensors supporting human performance and telemedicine. Veracity is demonstrated in action classification related to sport, specifically strikes in boxing and taekwondo. Four input models, formulated to be compatible with a broad range of classifiers, are introduced and two diverse classifiers, dynamic time warping (DTW) and convolutional neural networks (CNNs) are implemented in conjunction with the input models. Seven classification models fusing information at the input-level, output-level, and a combination of both are formulated. Action classification for 18 boxing punches and 24 taekwondo kicks demonstrate our fusion classifiers outperform the best DTW and CNN uni-axial classifiers. Furthermore, although DTW is ostensibly an ideal choice for human movements experiencing non-linear variations, our results demonstrate deep learning fusion classifiers outperform DTW. This is a novel finding given that CNNs are normally designed for multi-dimensional data and do not specifically compensate for non-linear variations within signal classes. The generalized formulation enables subject-specific movement classification in a feature-blind fashion with trivial computational expense for trained CNNs. A commercial boxing system, ‘Corner’, has been produced for real-world mass-market use based on this investigation providing a basis for future telemedicine translation. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

14 pages, 23794 KiB

Open AccessArticle

Videogrammetric Verification of Accuracy of Wearable Sensors Used in Kiteboarding

by Marián Marčiš, Marek Fraštia, Andrej Hideghéty and Peter Paulík

Sensors 2021, 21(24), 8353; https://doi.org/10.3390/s21248353 - 14 Dec 2021

Cited by 5 | Viewed by 2130

Abstract

Owing to the combination of windsurfing, snowboarding, wakeboarding, and paragliding, kiteboarding has gained an enormous number of fans worldwide. Enthusiasts compete to achieve the maximum height and length of jumps, speed, or total distance travelled. Several commercially available systems have been developed to [...] Read more.

Owing to the combination of windsurfing, snowboarding, wakeboarding, and paragliding, kiteboarding has gained an enormous number of fans worldwide. Enthusiasts compete to achieve the maximum height and length of jumps, speed, or total distance travelled. Several commercially available systems have been developed to measure these parameters. However, practice shows that the accuracy of the implemented sensors is debatable. In this study, we examined the accuracy of jump heights determined by sensors WOO2 and WOO3, and the Surfr app installed on an Apple iPhone SE 2016, compared to a combination of videogrammetric and geodetic measurements. These measurements were performed using four cameras located on the shore of the Danube River at Šamorín, Slovakia. The videogrammetrically-determined accuracy of jump heights was 0.03–0.09 m. This can be considered a reference for comparing the accuracy of off-the-shelf systems. The results show that all of the systems compared tend to overestimate jump heights, including an increase in error with increasing jump height. For jumps over 5 m, the deviations reached more than 20% of the actual jump height. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

9 pages, 2597 KiB

Open AccessCommunication

Step-to-Step Kinematic Validation between an Inertial Measurement Unit (IMU) 3D System, a Combined Laser+IMU System and Force Plates during a 50 M Sprint in a Cohort of Sprinters

by Roland van den Tillaar, Ryu Nagahara, Sam Gleadhill and Pedro Jiménez-Reyes

Sensors 2021, 21(19), 6560; https://doi.org/10.3390/s21196560 - 30 Sep 2021

Cited by 11 | Viewed by 2737

Abstract

The purpose was to compare step-by-step kinematics measured using force plates (criterion), an IMU only and a combined laser IMU system in well-trained sprinters. Fourteen male experienced sprinters performed a 50-m sprint. Step-by-step kinematics were measured by 50 force plates and compared with [...] Read more.

The purpose was to compare step-by-step kinematics measured using force plates (criterion), an IMU only and a combined laser IMU system in well-trained sprinters. Fourteen male experienced sprinters performed a 50-m sprint. Step-by-step kinematics were measured by 50 force plates and compared with an IMU-3D motion capture system and a combined laser+IMU system attached to each foot. Results showed that step kinematics (step velocity, length, contact and flight times) were different when measured with the IMU-3D system, compared with force plates, while the laser+IMU system, showed in general the same kinematics as measured with force plates without a systematic bias. Based upon the findings it can be concluded that the laser+IMU system is as accurate in measuring step-by-step kinematics as the force plate system. At the moment, the IMU-3D system is only accurate in measuring stride patterns (temporal parameters); it is not accurate enough to measure step lengths (spatial) and velocities due to the inaccuracies in step length, especially at high velocities. It is suggested that this laser+IMU system is valid and accurate, which can be used easily in training and competition to obtain step-by step kinematics and give direct feedback of this information during training and competition. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

18 pages, 2958 KiB

Open AccessArticle

The Effect of Cleat Position on Running Using Acceleration-Derived Data in the Context of Triathlon

by Stuart A. Evans, Daniel A. James, David Rowlands and James B. Lee

Sensors 2021, 21(17), 5899; https://doi.org/10.3390/s21175899 - 02 Sep 2021

Cited by 1 | Viewed by 4913

Abstract

Appropriate cycling cleat adjustment could improve triathlon performance in both cycling and running. Prior recommendations regarding cleat adjustment have comprised aligning the first metatarsal head above the pedal spindle or somewhat forward. However, contemporary research has questioned this approach in triathlons due to [...] Read more.

Appropriate cycling cleat adjustment could improve triathlon performance in both cycling and running. Prior recommendations regarding cleat adjustment have comprised aligning the first metatarsal head above the pedal spindle or somewhat forward. However, contemporary research has questioned this approach in triathlons due to the need to run immediately after cycling. Subsequently, moving the pedal cleat posteriorly could be more appropriate. This study evaluated the effectiveness of a triaxial accelerometer to determine acceleration magnitudes of the trunk in outdoor cycling in two different bicycle cleat positions and the consequential impact on trunk acceleration during running. Seven recreational triathletes performed a 20 km cycle and a 5 km run using their own triathlon bicycle complete with aerodynamic bars and gearing. Interpretation of data was evaluated based on cadence changes whilst triathletes cycled in an aerodynamic position in two cleat positions immediately followed by a self-paced overground run. The evaluation of accelerometer-derived data within a characteristic overground setting suggests a significant increase in total trunk acceleration magnitude during cycling with a posterior cleat with significant increases to longitudinal acceleration (p = 0.04) despite a small effect (d = 0.2) to the ratings of perceived exertion (RPE). Cycling with a posterior cleat significantly reduced longitudinal trunk acceleration in running and overall acceleration magnitudes (p < 0.0001) with a large effect size (d = 0.9) and a significant reduction in RPE (p = 0.02). In addition, running after cycling in a posterior cleat was faster compared to running after cycling in a standard cleat location. Practically, the magnitude of trunk acceleration during cycling in a posterior cleat position as well as running after posterior cleat cycling differed from that when cycling in the fore-aft position followed by running. Therefore, the notion that running varies after cycling is not merely an individual athlete’s perception, but a valid observation that can be modified when cleat position is altered. Training specifically with a posterior cleat in cycling might improve running performance when trunk accelerations are analysed. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

16 pages, 1244 KiB

Open AccessArticle

External Load and Muscle Activation Monitoring of NCAA Division I Basketball Team Using Smart Compression Shorts

by David N. Saucier, Samaneh Davarzani, Reuben F. Burch V, Harish Chander, Lesley Strawderman, Charles Freeman, Logan Ogden, Adam Petway, Aaron Duvall, Collin Crane and Anthony Piroli

Sensors 2021, 21(16), 5348; https://doi.org/10.3390/s21165348 - 08 Aug 2021

Cited by 6 | Viewed by 3126

Abstract

There is scarce research into the use of Strive Sense3 smart compression shorts to measure external load with accelerometry and muscle load (i.e., muscle activations) with surface electromyography in basketball. Sixteen external load and muscle load variables were measured from 15 National Collegiate [...] Read more.

There is scarce research into the use of Strive Sense3 smart compression shorts to measure external load with accelerometry and muscle load (i.e., muscle activations) with surface electromyography in basketball. Sixteen external load and muscle load variables were measured from 15 National Collegiate Athletic Association Division I men’s basketball players with 1137 session records. The data were analyzed for player positions of Centers (n = 4), Forwards (n = 4), and Guards (n = 7). Nonparametric bootstrapping was used to find significant differences between training and game sessions. Significant differences were found in all variables except Number of Jumps and all muscle load variables for Guards, and all variables except Muscle Load for Forwards. For Centers, the Average Speed, Average Max Speed, and Total Hamstring, Glute, Left, and Right Muscle variables were significantly different (p < 0.05). Principal component analysis was conducted on the external load variables. Most of the variance was explained within two principal components (70.4% in the worst case). Variable loadings of principal components for each position were similar during training but differed during games, especially for the Forward position. Measuring muscle activation provides additional information in which the demands of each playing position can be differentiated during training and competition. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

10 pages, 1044 KiB

Open AccessArticle

The Effects of Knee Flexion on Tennis Serve Performance of Intermediate Level Tennis Players

by Joana Ferreira Hornestam, Thales Rezende Souza, Fabrício Anício Magalhães, Mickäel Begon, Thiago Ribeiro Teles Santos and Sérgio Teixeixa Fonseca

Sensors 2021, 21(16), 5254; https://doi.org/10.3390/s21165254 - 04 Aug 2021

Cited by 6 | Viewed by 3763

Abstract

This study aimed to investigate the effects of knee flexion during the preparation phase of a serve on the tennis serve performance, using inertial sensors. Thirty-two junior tennis players were divided into two groups based on their maximum knee flexion during the preparation [...] Read more.

This study aimed to investigate the effects of knee flexion during the preparation phase of a serve on the tennis serve performance, using inertial sensors. Thirty-two junior tennis players were divided into two groups based on their maximum knee flexion during the preparation phase of serve: Smaller (SKF) and Greater (GKF) Knee Flexion. Their racket velocity, racket height, and knee extension velocity were compared during the tennis serve. Inertial sensors tracked participants’ shank, thigh, and racket motions while performing five first, flat, and valid serves. Knee flexion was analysed during the preparation phase of serve, knee extension velocity after this phase, racket velocity just before ball impact, and racket height at impact. Pre-impact racket velocity (mean difference [MD] = 3.33 km/h, p = 0.004) and the knee extension velocity (MD = 130.30 °/s, p = 0.012) were higher in the GKF than SKF; however, racket impact height was not different between groups (p = 0.236). This study’s findings support the importance of larger knee flexion during the preparation phase of serve-to-serve performance. This motion should be seen as a contributor to racket velocity. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

12 pages, 1276 KiB

Open AccessArticle

Automated Classification of Changes of Direction in Soccer Using Inertial Measurement Units

by Brian Reilly, Oliver Morgan, Gabriela Czanner and Mark A. Robinson

Sensors 2021, 21(14), 4625; https://doi.org/10.3390/s21144625 - 06 Jul 2021

Cited by 5 | Viewed by 4232

Abstract

Changes of direction (COD) are an important aspect of soccer match play. Understanding the physiological and biomechanical demands on players in games allows sports scientists to effectively train and rehabilitate soccer players. COD are conventionally recorded using manually annotated time-motion video analysis which [...] Read more.

Changes of direction (COD) are an important aspect of soccer match play. Understanding the physiological and biomechanical demands on players in games allows sports scientists to effectively train and rehabilitate soccer players. COD are conventionally recorded using manually annotated time-motion video analysis which is highly time consuming, so more time-efficient approaches are required. The aim was to develop an automated classification model based on multi-sensor player tracking device data to detect COD > 45°. Video analysis data and individual multi-sensor player tracking data (GPS, accelerometer, gyroscopic) for 23 academy-level soccer players were used. A novel ‘GPS-COD Angle’ variable was developed and used in model training; along with 24 GPS-derived, gyroscope and accelerometer variables. Video annotation was the ground truth indicator of occurrence of COD > 45°. The random forest classifier using the full set of features demonstrated the highest accuracy (AUROC = 0.957, 95% CI = 0.956–0.958, Sensitivity = 0.941, Specificity = 0.772. To balance sensitivity and specificity, model parameters were optimised resulting in a value of 0.889 for both metrics. Similarly high levels of accuracy were observed for random forest models trained using a reduced set of features, accelerometer-derived variables only, and gyroscope-derived variables only. These results point to the potential effectiveness of the novel methodology implemented in automatically identifying COD in soccer players. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

14 pages, 848 KiB

Open AccessArticle

A Comparison of Three Neural Network Approaches for Estimating Joint Angles and Moments from Inertial Measurement Units

by Marion Mundt, William R. Johnson, Wolfgang Potthast, Bernd Markert, Ajmal Mian and Jacqueline Alderson

Sensors 2021, 21(13), 4535; https://doi.org/10.3390/s21134535 - 01 Jul 2021

Cited by 43 | Viewed by 5353

Abstract

The application of artificial intelligence techniques to wearable sensor data may facilitate accurate analysis outside of controlled laboratory settings—the holy grail for gait clinicians and sports scientists looking to bridge the lab to field divide. Using these techniques, parameters that are difficult to [...] Read more.

The application of artificial intelligence techniques to wearable sensor data may facilitate accurate analysis outside of controlled laboratory settings—the holy grail for gait clinicians and sports scientists looking to bridge the lab to field divide. Using these techniques, parameters that are difficult to directly measure in-the-wild, may be predicted using surrogate lower resolution inputs. One example is the prediction of joint kinematics and kinetics based on inputs from inertial measurement unit (IMU) sensors. Despite increased research, there is a paucity of information examining the most suitable artificial neural network (ANN) for predicting gait kinematics and kinetics from IMUs. This paper compares the performance of three commonly employed ANNs used to predict gait kinematics and kinetics: multilayer perceptron (MLP); long short-term memory (LSTM); and convolutional neural networks (CNN). Overall high correlations between ground truth and predicted kinematic and kinetic data were found across all investigated ANNs. However, the optimal ANN should be based on the prediction task and the intended use-case application. For the prediction of joint angles, CNNs appear favourable, however these ANNs do not show an advantage over an MLP network for the prediction of joint moments. If real-time joint angle and joint moment prediction is desirable an LSTM network should be utilised. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

19 pages, 22983 KiB

Open AccessArticle

Combining Inertial Sensors and Machine Learning to Predict vGRF and Knee Biomechanics during a Double Limb Jump Landing Task

by Courtney R. Chaaban, Nathaniel T. Berry, Cortney Armitano-Lago, Adam W. Kiefer, Michael J. Mazzoleni and Darin A. Padua

Sensors 2021, 21(13), 4383; https://doi.org/10.3390/s21134383 - 26 Jun 2021

Cited by 12 | Viewed by 3195

Abstract

(1) Background: Biomechanics during landing tasks, such as the kinematics and kinetics of the knee, are altered following anterior cruciate ligament (ACL) injury and reconstruction. These variables are recommended to assess prior to clearance for return to sport, but clinicians lack access to [...] Read more.

(1) Background: Biomechanics during landing tasks, such as the kinematics and kinetics of the knee, are altered following anterior cruciate ligament (ACL) injury and reconstruction. These variables are recommended to assess prior to clearance for return to sport, but clinicians lack access to the current gold-standard laboratory-based assessment. Inertial sensors serve as a potential solution to provide a clinically feasible means to assess biomechanics and augment the return to sport testing. The purposes of this study were to (a) develop multi-sensor machine learning algorithms for predicting biomechanics and (b) quantify the accuracy of each algorithm. (2) Methods: 26 healthy young adults completed 8 trials of a double limb jump landing task. Peak vertical ground reaction force, peak knee flexion angle, peak knee extension moment, and peak sagittal knee power absorption were assessed using 3D motion capture and force plates. Shank- and thigh- mounted inertial sensors were used to collect data concurrently. Inertial data were submitted as inputs to single- and multiple- feature linear regressions to predict biomechanical variables in each limb. (3) Results: Multiple-feature models, particularly when an accelerometer and gyroscope were used together, were valid predictors of biomechanics (R² = 0.68–0.94, normalized root mean square error = 4.6–10.2%). Single-feature models had decreased performance (R² = 0.16–0.60, normalized root mean square error = 10.0–16.2%). (4) Conclusions: The combination of inertial sensors and machine learning provides a valid prediction of biomechanics during a double limb landing task. This is a feasible solution to assess biomechanics for both clinical and real-world settings outside the traditional biomechanics laboratory. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

15 pages, 4892 KiB

Open AccessArticle

The Protraction and Retraction Angles of Horse Limbs: An Estimation during Trotting Using Inertial Sensors

by Marie Sapone, Pauline Martin, Khalil Ben Mansour, Henry Chateau and Frédéric Marin

Sensors 2021, 21(11), 3792; https://doi.org/10.3390/s21113792 - 30 May 2021

Cited by 5 | Viewed by 4219

Abstract

The protraction and retraction angles of horse limbs are important in the analysis of horse locomotion. This study explored two methods from an IMU positioned on the canon bone of eight horses to estimate these angles. Each method was based on a hypothesis [...] Read more.

The protraction and retraction angles of horse limbs are important in the analysis of horse locomotion. This study explored two methods from an IMU positioned on the canon bone of eight horses to estimate these angles. Each method was based on a hypothesis in order to define the moment corresponding with the verticality of the canon bone: (i) the canon bone is in a vertical position at 50% of the stance phase or (ii) the verticality of the canon bone corresponds with the moment when the horse’s withers reach their lowest point. The measurements were carried out on a treadmill at a trot and compared with a standard gold method based on motion capture. For the measurement of the maximum protraction and retraction angles, method (i) had average biases (0.7° and 1.7°) less than method (ii) (−1.3° and 3.7°). For the measurement of the protraction and retraction angles during the stance phase, method (i) had average biases (4.1° and −3.3°) higher to method (ii) (2.1° and −1.3°). This study investigated the pros and cons of a generic method (i) vs. a specific method (ii) to determine the protraction and retraction angles of horse limbs by a single IMU. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

11 pages, 1463 KiB

Open AccessCommunication

Ankle Muscle Activations during Different Foot-Strike Patterns in Running

by Jian-Zhi Lin, Wen-Yu Chiu, Wei-Hsun Tai, Yu-Xiang Hong and Chung-Yu Chen

Sensors 2021, 21(10), 3422; https://doi.org/10.3390/s21103422 - 14 May 2021

Cited by 4 | Viewed by 2625

Abstract

This study analysed the landing performance and muscle activity of athletes in forefoot strike (FFS) and rearfoot strike (RFS) patterns. Ten male college participants were asked to perform two foot strikes patterns, each at a running speed of 6 km/h. Three inertial sensors [...] Read more.

This study analysed the landing performance and muscle activity of athletes in forefoot strike (FFS) and rearfoot strike (RFS) patterns. Ten male college participants were asked to perform two foot strikes patterns, each at a running speed of 6 km/h. Three inertial sensors and five EMG sensors as well as one 24 G accelerometer were synchronised to acquire joint kinematics parameters as well as muscle activation, respectively. In both the FFS and RFS patterns, according to the intraclass correlation coefficient, excellent reliability was found for landing performance and muscle activation. Paired t tests indicated significantly higher ankle plantar flexion in the FFS pattern. Moreover, biceps femoris (BF) and gastrocnemius medialis (GM) activation increased in the pre-stance phase of the FFS compared with that of RFS. The FFS pattern had significantly decreased tibialis anterior (TA) muscle activity compared with the RFS pattern during the pre-stance phase. The results demonstrated that the ankle strategy focused on controlling the foot strike pattern. The influence of the FFS pattern on muscle activity likely indicates that an athlete can increase both BF and GM muscles activity. Altered landing strategy in cases of FFS pattern may contribute both to the running efficiency and muscle activation of the lower extremity. Therefore, neuromuscular training and education are required to enable activation in dynamic running tasks. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

14 pages, 2454 KiB

Open AccessArticle

A Novel Accelerometry-Based Metric to Improve Estimation of Whole-Body Mechanical Load

by Enzo Hollville, Antoine Couturier, Gaël Guilhem and Giuseppe Rabita

Sensors 2021, 21(10), 3398; https://doi.org/10.3390/s21103398 - 13 May 2021

Cited by 9 | Viewed by 2545

Abstract

While the Player Load is a widely-used parameter for physical demand quantification using wearable accelerometers, its calculation is subjected to potential errors related to rotational changes of the reference frame. The aims of this study were (i) to assess the concurrent validity of [...] Read more.

While the Player Load is a widely-used parameter for physical demand quantification using wearable accelerometers, its calculation is subjected to potential errors related to rotational changes of the reference frame. The aims of this study were (i) to assess the concurrent validity of accelerometry-based Player Load against force plates; (ii) to validate a novel metric, the Accel’Rate overcoming this theoretical issue. Twenty-one recreational athlete males instrumented with two triaxial accelerometers positioned at the upper and lower back performed running-based locomotor movements at low and high intensity over six in-series force plates. We examined the validity of the Player Load and the Accel’Rate by using force plates. Standard error of the estimate was small to moderate for all tested conditions (Player Load: 0.45 to 0.87; Accel’Rate: 0.25 to 0.95). Accel’Rate displayed trivial to small mean biases (−1.0 to 6.1 a.u.) while the Player Load displayed systematic very large to extremely large mean biases (17.1 to 226.0 a.u.). These findings demonstrate a better concurrent validity of the Accel’Rate compared to the Player Load. This metric could be used to improve the estimation of whole-body mechanical load, easily accessible in sport training and competition settings. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

10 pages, 684 KiB

Open AccessArticle

Punch Trackers: Correct Recognition Depends on Punch Type and Training Experience

by Dan Omcirk, Tomas Vetrovsky, Jan Padecky, Sophie Vanbelle, Jan Malecek and James J. Tufano

Sensors 2021, 21(9), 2968; https://doi.org/10.3390/s21092968 - 23 Apr 2021

Cited by 10 | Viewed by 3110

Abstract

To determine the ability of different punch trackers (PT) (Corner (CPT), Everlast (EPT), and Hykso (HPT)) to recognize specific punch types (lead and rear straight punches, lead and rear hooks, and lead and rear uppercuts) thrown by trained (TR, n = 10) and [...] Read more.

To determine the ability of different punch trackers (PT) (Corner (CPT), Everlast (EPT), and Hykso (HPT)) to recognize specific punch types (lead and rear straight punches, lead and rear hooks, and lead and rear uppercuts) thrown by trained (TR, n = 10) and untrained punchers (UNTR, n = 11), subjects performed different punch combinations, and PT data were compared to data from video recordings to determine how well each PT recognized the punches that were actually thrown. Descriptive statistics and multilevel modelling were used to analyze the data. The CPT, EPT and HPT detected punches more accurately in TR than UNTR, evidenced by a lower percentage error in TR (p = 0.007). The CPT, EPT, and HPT detected straight punches better than uppercuts and hooks, with a lower percentage error for straight punches (p < 0.001). The recognition of punches with CPT and HPT depended on punch order, with earlier punches in a sequence recognized better. The same may or may not have occurred with EPT, but EPT does not allow for data to be exported, meaning the order of individual punches could not be analyzed. The CPT and HPT both seem to be viable options for tracking punch count and punch type in TR and UNTR. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

18 pages, 19369 KiB

Open AccessArticle

Can Markerless Pose Estimation Algorithms Estimate 3D Mass Centre Positions and Velocities during Linear Sprinting Activities?

by Laurie Needham, Murray Evans, Darren P. Cosker and Steffi L. Colyer

Sensors 2021, 21(8), 2889; https://doi.org/10.3390/s21082889 - 20 Apr 2021

Cited by 16 | Viewed by 4234

Abstract

The ability to accurately and non-invasively measure 3D mass centre positions and their derivatives can provide rich insight into the physical demands of sports training and competition. This study examines a method for non-invasively measuring mass centre velocities using markerless human pose estimation [...] Read more.

The ability to accurately and non-invasively measure 3D mass centre positions and their derivatives can provide rich insight into the physical demands of sports training and competition. This study examines a method for non-invasively measuring mass centre velocities using markerless human pose estimation and Kalman smoothing. Marker (Qualysis) and markerless (OpenPose) motion capture data were captured synchronously for sprinting and skeleton push starts. Mass centre positions and velocities derived from raw markerless pose estimation data contained large errors for both sprinting and skeleton pushing (mean ± SD = 0.127 ± 0.943 and −0.197 ± 1.549 m·s⁻¹, respectively). Signal processing methods such as Kalman smoothing substantially reduced the mean error (±SD) in horizontal mass centre velocities (0.041 ± 0.257 m·s⁻¹) during sprinting but the precision remained poor. Applying pose estimation to activities which exhibit unusual body poses (e.g., skeleton pushing) appears to elicit more erroneous results due to poor performance of the pose estimation algorithm. Researchers and practitioners should apply these methods with caution to activities beyond sprinting as pose estimation algorithms may not generalise well to the activity of interest. Retraining the model using activity specific data to produce more specialised networks is therefore recommended. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

7 pages, 486 KiB

Open AccessCommunication

Can Machine Learning with IMUs Be Used to Detect Different Throws and Estimate Ball Velocity in Team Handball?

by Roland van den Tillaar, Shruti Bhandurge and Tom Stewart

Sensors 2021, 21(7), 2288; https://doi.org/10.3390/s21072288 - 25 Mar 2021

Cited by 8 | Viewed by 2734

Abstract

Injuries in handball are common due to the repetitive demands of overhead throws at high velocities. Monitoring workload is crucial for understanding these demands and improving injury-prevention strategies. However, in handball, it is challenging to monitor throwing workload due to the difficulty of [...] Read more.

Injuries in handball are common due to the repetitive demands of overhead throws at high velocities. Monitoring workload is crucial for understanding these demands and improving injury-prevention strategies. However, in handball, it is challenging to monitor throwing workload due to the difficulty of counting the number, intensity, and type of throws during training and competition. The aim of this study was to investigate if an inertial measurement unit (IMU) and machine learning (ML) techniques could be used to detect different types of team handball throws and predict ball velocity. Seventeen players performed several throws with different wind-up (circular and whip-like) and approach types (standing, running, and jumping) while wearing an IMU on their wrist. Ball velocity was measured using a radar gun. ML models predicted peak ball velocity with an error of 1.10 m/s and classified approach type and throw type with 80–87% accuracy. Using IMUs and ML models may offer a practical and automated method for quantifying throw counts and classifying the throw and approach types adopted by handball players. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

17 pages, 6528 KiB

Open AccessArticle

Tracking Quantitative Characteristics of Cutting Maneuvers with Wearable Movement Sensors during Competitive Women’s Ultimate Frisbee Games

by Paul R. Slaughter and Peter G. Adamczyk

Sensors 2020, 20(22), 6508; https://doi.org/10.3390/s20226508 - 14 Nov 2020

Cited by 10 | Viewed by 3911

Abstract

(1) Ultimate frisbee involves frequent cutting motions, which have a high risk of anterior cruciate ligament (ACL) injury, especially for female players. This study investigated the in-game cutting maneuvers performed by female ultimate frisbee athletes to understand the movements that could put them [...] Read more.

(1) Ultimate frisbee involves frequent cutting motions, which have a high risk of anterior cruciate ligament (ACL) injury, especially for female players. This study investigated the in-game cutting maneuvers performed by female ultimate frisbee athletes to understand the movements that could put them at risk of ACL injury. (2) Lower-body kinematics and movement around the field were reconstructed from wearable lower-body inertial sensors worn by 12 female players during 16 league-sanctioned ultimate frisbee games. (3) 422 cuts were identified from speed and direction change criteria. The mean cut had approach speed of 3.4 m/s, approach acceleration of 3.1 m/s², cut angle of 94 degrees, and ground-contact knee flexion of 34 degrees. Shallow cuts from 30 to 90 degrees were most common. Speed and acceleration did not change based on cut angle. Players on more competitive teams had higher speed and acceleration and reduced knee flexion during cutting. (4) This study demonstrates that a lower-body set of wearable inertial sensors can successfully track an athlete’s motion during real games, producing detailed biomechanical metrics of behavior and performance. These in-game measurements can be used to specify controlled cutting movements in future laboratory studies. These studies should prioritize higher-level players since they may exhibit higher-risk cutting behavior. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Graphical abstract

15 pages, 2637 KiB

Open AccessArticle

Fusing Accelerometry with Videography to Monitor the Effect of Fatigue on Punching Performance in Elite Boxers

by Nicos Haralabidis, David John Saxby, Claudio Pizzolato, Laurie Needham, Dario Cazzola and Clare Minahan

Sensors 2020, 20(20), 5749; https://doi.org/10.3390/s20205749 - 10 Oct 2020

Cited by 12 | Viewed by 4290

Abstract

Wearable sensors and motion capture technology are accepted instruments to measure spatiotemporal variables during punching performance and to study the externally observable effects of fatigue. This study aimed to develop a computational framework enabling three-dimensional inverse dynamics analysis through the tracking of punching [...] Read more.

Wearable sensors and motion capture technology are accepted instruments to measure spatiotemporal variables during punching performance and to study the externally observable effects of fatigue. This study aimed to develop a computational framework enabling three-dimensional inverse dynamics analysis through the tracking of punching kinematics obtained from inertial measurement units and uniplanar videography. The framework was applied to six elite male boxers performing a boxing-specific punch fatigue protocol. OpenPose was used to label left side upper-limb landmarks from which sagittal plane kinematics were computed. Custom-made inertial measurement units were embedded into the boxing gloves, and three-dimensional punch accelerations were analyzed using statistical parametric mapping to evaluate the effects of both fatigue and laterality. Tracking simulations of a sub-set of left-handed punches were formulated as optimal control problems and converted to nonlinear programming problems for solution with a trapezoid collocation method. The laterality analysis revealed the dominant side fatigued more than the non-dominant, while tracking simulations revealed shoulder abduction and elevation moments increased across the fatigue protocol. In future, such advanced simulation and analysis could be performed in ecologically valid contexts, whereby multiple inertial measurement units and video cameras might be used to model a more complete set of dynamics. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

Other

Jump to: Research

8 pages, 726 KiB

Open AccessLetter

Estimating Throwing Speed in Handball Using a Wearable Device

by Sebastian D. Skejø, Jesper Bencke, Merete Møller and Henrik Sørensen

Sensors 2020, 20(17), 4925; https://doi.org/10.3390/s20174925 - 31 Aug 2020

Cited by 9 | Viewed by 3760

Abstract

Throwing speed is likely a key determinant of shoulder-specific load. However, it is difficult to estimate the speed of throws in handball in field-based settings with many players due to limitations in current technology. Therefore, the purpose of this study was to develop [...] Read more.

Throwing speed is likely a key determinant of shoulder-specific load. However, it is difficult to estimate the speed of throws in handball in field-based settings with many players due to limitations in current technology. Therefore, the purpose of this study was to develop a novel method to estimate throwing speed in handball using a low-cost accelerometer-based device. Nineteen experienced handball players each performed 25 throws of varying types while we measured the acceleration of the wrist using the accelerometer and the throwing speed using 3D motion capture. Using cross-validation, we developed four prediction models using combinations of the logarithm of the peak total acceleration, sex and throwing type as the predictor and the throwing speed as the outcome. We found that all models were well-calibrated (mean calibration of all models: 0.0 m/s, calibration slope of all models: 1.00) and precise (R2 = 0.71–0.86, mean absolute error = 1.30–1.82 m/s). We conclude that the developed method provides practitioners and researchers with a feasible and cheap method to estimate throwing speed in handball from segments of wrist acceleration signals containing only a single throw. Full article

(This article belongs to the Special Issue Sensors in Sports Biomechanics)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Sensors in Sports Biomechanics

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Published Papers (19 papers)

Research

Other

Further Information

Guidelines

MDPI Initiatives

Follow MDPI