Inertial Sensor Technology for Elite Swimming Performance Analysis: A Systematic Review

Profiling biomechanical abilities during sprint front-crawl swimming using IMU and functional clustering of variabilities

This study aims to profile biomechanical abilities during sprint front crawl by identifying technical stroke characteristics, in light of performance level. Ninety-one recreational to world-class swimmers equipped with a sacrum-worn IMU performed 25 m all-out. Intra and inter-cyclic 3D kinematical variabilities were clustered using a functional double partition model. Clusters were analysed according to (1) swimming technique using continuous visualisation and discrete features (standard deviation and jerk cost) and (2) performance regarding speed and competition calibre using respectively one-way ANOVA and Chi-squared test as well as Gamma statistics. Swimmers displayed specific technical profiles of intra-cyclic (smoothy and jerky) and inter-cyclic stroke regulation (low, moderate and high repeatability) significantly discriminated by speed (p < 0.001, η2 = 0.62) and performance calibre (p < 0.001, V = 0.53). We showed that combining high levels of both kinds of variability (jerky + low repeatability) are associated with highest speed (1.86 ± 0.12 m/s) and competition calibre (ℽ = 0.75, p < 0.001). It highlights the crucial importance of variabilities combination. Technical skills might be driven by a specific alignment of stroke pattern and its associated dispersion according to the task constraints. This data-driven approach can assist eyes-based technical evaluation. Targeting the development of an explosive swimming style with a high level of body stability should be considered during training of sprinters.

Kinematic characteristics of elbow joint range of motion in elite Chinese freestyle swimmers with elbow pain during dry-land simulations of swimming strokes

The aim of this study was to obtain quantitative data on elbow joint ROM in elite freestyle swimmers with EP in China. Of the 50 elite freestyle swimmers recruited, 41 completed all measurements during dry-land swimming stroke simulations. Elbow joint angle, velocity, and acceleration were measured using inertial measurement units. The RMSE/D was calculated to determine the elbow joint ROM deviation. Joint angle (3.33  ∘ -42.96  ∘ ), angular velocity (-364.15 to 245.69  ∘ / s ), and angular acceleration (-7051.80 to 1465.35  ∘ / s 2 ) were significantly different between the critical pain and healthy. The probability distributions of joint angle (15.47  ∘ ±14.54  ∘ ), angular velocity (2.41  ∘ ±111.06  ∘ / s ), and angular acceleration (1.93 ± 2222.6  ∘ / s 2 ) in the slight pain group were significantly different betweenhealthy and critical pain. The RMSE/D distributions of angular velocity (28.3%) and acceleration (21.48%) in the critical pain deviated from the healthy. The peak value-RMSE/D matrix model obtained proved that elbow ROM significantly differed between the elite freestyle swimmers with EP and the healthy. Angular velocity and acceleration indicate the weakness and negative influence of kinematics on patients with EP. Thus, Potential solutions are to constantly optimise freestyle swimming techniques and strengthen the arm muscles.

BackMov: Individualized Motion Capture-Based Test to Assess Low Back Pain Mobility Recovery after Treatment

Low back pain (LBP) is a common issue that negatively affects a person's quality of life and imposes substantial healthcare expenses. In this study, we introduce the (Back-pain Movement) BackMov test, using inertial motion capture (MoCap) to assess lumbar movement changes in LBP patients. The test includes flexion-extension, rotation, and lateralization movements focused on the lumbar spine. To validate its reproducibility, we conducted a test-retest involving 37 healthy volunteers, yielding results to build a minimal detectable change (MDC) graph map that would allow us to see if changes in certain variables of LBP patients are significant in relation to their recovery. Subsequently, we evaluated its applicability by having 30 LBP patients perform the movement's test before and after treatment (15 received deep oscillation therapy; 15 underwent conventional therapy) and compared the outcomes with a specialist's evaluations. The test-retest results demonstrated high reproducibility, especially in variables such as range of motion, flexion and extension ranges, as well as velocities of lumbar movements, which stand as the more important variables that are correlated with LBP disability, thus changes in them may be important for patient recovery. Among the 30 patients, the specialist's evaluations were confirmed using a low-back-specific Short Form (SF)-36 Physical Functioning scale, and agreement was observed, in which all patients improved their well-being after both treatments. The results from the specialist analysis coincided with changes exceeding MDC values in the expected variables. In conclusion, the BackMov test offers sensitive variables for tracking mobility recovery from LBP, enabling objective assessments of improvement. This test has the potential to enhance decision-making and personalized patient monitoring in LBP management.

Validation of Automatically Quantified Swim Stroke Mechanics Using an Inertial Measurement Unit in Paralympic Athletes

Biomechanics and training load monitoring are important for performance evaluation and injury prevention in elite swimming. Monitoring of performance and swim stroke parameters is possible with inertial measurement units (IMU) but has not been validated in para-swimmers. The purpose of this study was to validate a single IMU-based system to accurately estimate pool-swam lap time, stroke count (SC), stroke duration, instantaneous stroke rate (ISR), and distance per stroke (DPS). Eight Paralympic athletes completed 4 × 50 m swims with an IMU worn on the sacrum. Strokes cycles were identified using a zero-crossing algorithm on the medio-lateral (freestyle and backstroke) or forward-backward (butterfly and breaststroke) instantaneous velocity data. Video-derived metrics were estimated using Dartfish and Kinovea. Agreement analyses, including Bland-Altman and Intraclass Correlation Coefficient (ICC), were performed on all outcome variables. SC Bland-Altman bias was 0.13 strokes, and ICC was 0.97. ISR Bland-Altman biases were within 1.5 strokes/min, and ICCs ranged from 0.26 to 0.96. DPS Bland-Altman biases were within 0.20 m, and ICCs ranged from 0.39 to 0.93. A single-IMU system can provide highly valid performance and swim stroke monitoring data for elite para-swimmers for the majority of strokes, with the exception of backstroke. Future work should improve bilateral stroke detection algorithms in this population.

Coordination and stroking parameters in the four swimming techniques: a narrative review

Swimming performances are multifactorial and primarily include anthropometric, hydrodynamic, bioenergetic and biomechanical factors whose contributions depend on age, gender, swimming distance and swimming stroke. An integrative and multivariate approach to swimming captures the complexity and various pathways of performance, but swimming technique is generally examined through such parameters as stroke index, propelling efficiency, stroke length and stroke rate. The first originality of our narrative review is to present the state of art of the methods to collect and measure inter-limb coordination in the four swimming techniques, with a particular focus on the effect of skill. The second part provides readers with an overview of the current findings on the main factors that influence inter-limb coordination (i.e., swimming speed, drag and the manipulation of stroke rate) following a physical approach.

The metronome-based methodology to monitor the stroke length changes in trained swimmers

The aim of our study was to develop a methodology that uses the metronome to constrain the swimmers' stroke rate with the aim to monitor changes in stroke length (SL) during two different periods of the season. Thirteen young trained swimmers (15.7 ± 1.7 y) performed three 50 m front crawl time trials during pre-season (PRE) and after 2 months, during the in-season period (IN). They were asked: (I) to swim at their maximum intensity (NO-MET condition); (II) to synchronize their stroke with a metronome beat set to their preferred intra-stroke-interval (ISI) (100% condition, corresponding to 48 ± 0.7 cycles/min); (III) to synchronize their stroke with a metronome beat set at 5% higher than their preferred ISI (95% condition, corresponding to 51 ± 0.8 cycles/min). The outcome parameters used to evaluate the performance were ISI, SL and total time of 50 m (TT). In NO-MET condition, results showed that TT in IN improved with respect to PRE, but no changes in ISI and SL. In 100% condition, no differences were obtained between the imposed and the performed ISI, whilst in 95% condition, the performed ISI was lower than the metronome ISI, and lower than that in 100% condition. At last, when using the metronome, SL was higher during IN compared to PRE and SL was lower in the 95% condition compared to the 100% condition. Results indicate that the use of the metronome successfully allowed monitoring changes in SL during different periods of the season. This methodology provides valuable information to coaches and athletes to enhance their performance throughout the season.

Profiles of stroke regulations discriminate between finishing positions during international open water races

This study aims to identify stroke regulation profiles and tipping-points in stroke regulation timing during international open water races according to performance level. Twelve elite or world-class swimmers were analysed during 18 international races. Stroke rate and jerk cost were computed cycle-to-cycle using an Inertial Measurement Unit and regulations profiles fitted using polynomials. We performed two-ways mixed-ANOVA to compare stroke kinematics among race segments and performance groups (G1 -fastest- to G3 -slowest-). Swimmers displayed specific regulation profiles (i.e., J-shape with end-spurt, J-shape without end-spurt and reverse L-shape for stroke rate and U-shape, reverse J-shape and reverse L-shape for jerk cost, for respectively G1, G2 and G3) with significant effect of race segment on stroke kinematics for G1 and G2. We highlighted tipping-points in stroke regulations profiles (TP1 and TP2) at respectively 30% and 75% of the race with greater magnitude in G1 than G2. TP1 reflects the end of a stroke economy period (0-30%) and TP2 the end of a progressive increase in stroke kinematics (30-75%) towards end-spurt (75-100%). Open water races follow a high-grading dynamics requiring biomechanical regulations along the race. Targeting stroke rate reserve and management of stroke smoothness should be considered during training of open water swimmers.

Augmented-reality swim goggles accurately and reliably measure swim performance metrics in recreational swimmers

Background

Swimmers commonly access performance metrics such as lap splits, distance, and pacing information between work bouts while they rest. Recently, a new category of tracking devices for swimming was introduced with the FORM Smart Swim Goggles (FORM Goggles). The goggles have a built-in see-through display and are capable of tracking and displaying distance, time splits, stroke, and pace metrics in real time using machine learning and augmented reality through a heads-up display. The purpose of this study was to assess the validity and reliability of the FORM Goggles compared with video analysis for stroke type, pool length count, pool length time, stroke rate, and stroke count in recreational swimmers and triathletes.

Method

A total of 36 participants performed mixed swimming intervals in a 25-m pool across two identical 900-m swim sessions performed at comparable intensities with 1 week interval. The participants wore FORM Goggles during their swims, which detected the following five swim metrics: stroke type, pool length time, pool length count, stroke count, and stroke rate. Four video cameras were positioned on the pool edges to capture ground truth video footage, which was then manually labeled by three trained individuals. Mean (SD) differences between FORM Goggles and ground truth were calculated for the selected metrics for both sessions. The absolute mean difference and mean absolute percentage error were used to assess the differences of the FORM Goggles relative to ground truth. The test-retest reliability of the goggles was assessed using both relative and absolute reliability metrics.

Results

Compared with video analysis, the FORM Goggles identified the correct stroke type at a rate of 99.7% (N = 2,354 pool lengths, p < 0.001), pool length count accuracy of 99.8%, and mean differences (FORM Goggles-ground truth) for pool length time: -0.10 s (1.49); stroke count: -0.63 (1.82); and stroke rate: 0.19 strokes/min (3.23). The test-retest intra-class correlation coefficient (ICC) values between the two test days were 0.793 for pool length time, 0.797 for stroke count, and 0.883 for stroke rate. Overall, for pool length time, the residuals were within ±1.0s for 65.3% of the total pool lengths, for stroke count within ±1 stroke for 62.6% of the total pool lengths, and for stroke rate within ±2 strokes/min for 66.40% of the total pool lengths.

Conclusion

The FORM Goggles were found valid and reliable for the tracking of pool length time, pool length count, stroke count, stroke rate, and stroke type during freestyle, backstroke, and breaststroke swimming in recreational swimmers and triathletes when compared with video analysis. This opens perspectives for receiving real-time information on performance metrics during swimming.

Device for Dual Ultrasound and Dry Needling Trigger Points Treatment

Ultrasound is a well-known tool to produce thermal and non-thermal effects on cells and tissues. These effects require an appropriate application of ultrasound in terms of localization and acoustic energy delivered. This article describes a new device that combines ultrasound and dry needling treatments. The non-thermal effects of ultrasound should locally amplify the needle's effects. The ultrasound transducer can mechanically rotate in 3D space to align itself in the direction of the needle. The transducer electronically focuses the acoustic pressure automatically on the needle tip and its surroundings. A computer, using graphical interface software, controls the angulation of the array and the focus position.

Digital Devices for Assessing Motor Functions in Mobility-Impaired and Healthy Populations: Systematic Literature Review

Background

With the advent of smart sensing technology, mobile and wearable devices can provide continuous and objective monitoring and assessment of motor function outcomes.

Objective

We aimed to describe the existing scientific literature on wearable and mobile technologies that are being used or tested for assessing motor functions in mobility-impaired and healthy adults and to evaluate the degree to which these devices provide clinically valid measures of motor function in these populations.

Methods

A systematic literature review was conducted by searching Embase, MEDLINE, CENTRAL (January 1, 2015, to June 24, 2020), the United States and European Union clinical trial registries, and the United States Food and Drug Administration website using predefined study selection criteria. Study selection, data extraction, and quality assessment were performed by 2 independent reviewers.

Results

A total of 91 publications representing 87 unique studies were included. The most represented clinical conditions were Parkinson disease (n=51 studies), followed by stroke (n=5), Huntington disease (n=5), and multiple sclerosis (n=2). A total of 42 motion-detecting devices were identified, and the majority (n=27, 64%) were created for the purpose of health care–related data collection, although approximately 25% were personal electronic devices (eg, smartphones and watches) and 11% were entertainment consoles (eg, Microsoft Kinect or Xbox and Nintendo Wii). The primary motion outcomes were related to gait (n=30), gross motor movements (n=25), and fine motor movements (n=23). As a group, sensor-derived motion data showed a mean sensitivity of 0.83 (SD 7.27), a mean specificity of 0.84 (SD 15.40), a mean accuracy of 0.90 (SD 5.87) in discriminating between diseased individuals and healthy controls, and a mean Pearson r validity coefficient of 0.52 (SD 0.22) relative to clinical measures. We did not find significant differences in the degree of validity between in-laboratory and at-home sensor-based assessments nor between device class (ie, health care–related device, personal electronic devices, and entertainment consoles).

Conclusions

Sensor-derived motion data can be leveraged to classify and quantify disease status for a variety of neurological conditions. However, most of the recent research on digital clinical measures is derived from proof-of-concept studies with considerable variation in methodological approaches, and much of the reviewed literature has focused on clinical validation, with less than one-quarter of the studies performing analytical validation. Overall, future research is crucially needed to further consolidate that sensor-derived motion data may lead to the development of robust and transformative digital measurements intended to predict, diagnose, and quantify neurological disease state and its longitudinal change.

Reliability of using a pressure sensor system to measure in-water force in young competitive swimmers

The aim of this study was to analyze the reliability of using a differential pressure system to measure in-water force in young competitive swimmers. Ten boys and five girls (12.38 ± 0.48 years, 49.13 ± 6.82 kg, 159.71 ± 7.99 cm) were randomly assigned to perform two maximum bouts of 25 m front crawl on different days (trial one, T1; trial two, T2), one week apart. A differential pressure system composed of two hand sensors (Aquanex System, v.4.1, Model DU2, Type A, Swimming Technology Research, Richmond, VA, United States) was used to measure the peak (RF_PEAK) and the mean (RF_MEAN) resultant force of the dominant and non-dominant hands (in Newton, N). Reliability was analyzed by computing the intraclass correlation coefficient (ICC), typical error (TE), smallest worthwhile change (SWC), coefficient of variation (CV%), standard error of measurement (SEM), and the minimal detectable change (MDC). Bland–Altman plots with 95% limits of agreement were also analyzed. The results showed no differences between T1 and T2 in all variables (p > 0.05). The ICC showed “excellent” reliability (ICC > 0.90) for the RF_PEAK and RF_MEAN in both hands. The CV% was rated as “good” (<5%) and TE was smaller than SWC in all variables. The Bland-Altman plots showed high reliability with a small bias (RF_PEAK dominant, -0.29 N; RF_PEAK non-dominant, -0.83 N; RF_MEAN dominant, 0.03 N; RF_MEAN non-dominant, 0.50 N). The pressure sensor system (Aquanex System) seems to be a reliable device for measuring the hand resultant force during front crawl in young swimmers and can be used to monitor the changes over time.

Novel Method for Estimating Propulsive Force Generated by Swimmers' Hands Using Inertial Measurement Units and Pressure Sensors

Propulsive force is a determinant of swimming performance. Several methods have been proposed to estimate the propulsive force in human swimming; however, their practical use in coaching is limited. Herein, we propose a novel method for estimating the propulsive force generated by swimmers' hands using an inertial measurement unit (IMU) and pressure sensors. In Experiment 1, we use a hand model to examine the effect of a hand-mounted IMU on pressure around the hand model at several flow velocities and water flow directions. In Experiment 2, we compare the propulsive force estimated using the IMU and pressure sensors (FIMU) via an underwater motion-capture system and pressure sensors (FMocap). Five swimmers had markers, pressure sensors, and IMUs attached to their hands and performed front crawl swimming for 25 m twice at each of nine different swimming speeds. The results show that the hand-mounted IMU affects the resultant force; however, the effect of the hand-mounted IMU varies with the flow direction. The mean values of FMocap and FIMU are similar (19.59 ± 7.66 N and 19.36 ± 7.86 N, respectively; intraclass correlation coefficient_(2,1) = 0.966), and their waveforms are similar (coefficient of multiple correlation = 0.99). These results indicate that the IMU can estimate the same level of propulsive force as an underwater motion-capture system.

Monitoring weekly progress of front crawl swimmers using IMU-based performance evaluation goal metrics

Technical evaluation of swimming performance is an essential factor in preparing elite swimmers for their competitions. Inertial measurement units (IMUs) have attracted much attention recently because they can provide coaches with a detailed analysis of swimmers’ performance during training. A coach can obtain a quantitative and objective evaluation from IMU. The purpose of this study was to validate the use of a new phase-based performance assessment with a single IMU worn on the sacrum during training sessions. Sixteen competitive swimmers performed five one-way front crawl trials at their maximum speed wearing an IMU on the sacrum. The coach recorded the lap time for each trial, as it remains the gold standard for swimmer’s performance in competition. The measurement was carried out once a week for 10 consecutive weeks to monitor the improvement in the swimmers’ performance. Meaningful progress was defined as a time decrease of at least 0.5 s over a 25 m lap. Using validated algorithms, we estimated five goal metrics from the IMU signals representing the swimmer’s performance in the swimming phases (wall push-off, glide, stroke preparation, free-swimming) and in the entire lap. The results showed that the goal metrics for free-swimming phase and the entire lap predicted the swimmer’s progress well (e.g., accuracy, precision, sensitivity, and specificity of 0.91, 0.89, 0.94, and 0.95 for the lap goal metric, respectively). As the goal metrics for initial phases (wall push-off, glide, stroke preparation) achieved high precision and specificity (≥0.79) in progress detection, the coach can use them for swimmers with satisfactory free-swimming phase performance and make further improvements in initial phases. Changes in the values of the goal metrics have been shown to be correlated with changes in lap time when there is meaningful progress. The results of this study show that goal metrics provided by the phase-based performance evaluation with a single IMU can help monitoring swimming progress. Average velocity of the lap can replace traditional lap time measurement, while phase-based goal metrics provide more information about the swimmer’s performance in each phase. This evaluation can help the coach quantitatively monitor the swimmer’s performance and train them more efficiently.

Automatic Swimming Activity Recognition and Lap Time Assessment Based on a Single IMU: A Deep Learning Approach

This study presents a deep learning model devoted to the analysis of swimming using a single Inertial Measurement Unit (IMU) attached to the sacrum. Gyroscope and accelerometer data were collected from 35 swimmers with various expertise levels during a protocol including the four swimming techniques. The proposed methodology took high inter- and intra-swimmer variability into account and was set up for the purpose of predicting eight swimming classes (the four swimming techniques, rest, wallpush, underwater, and turns) at four swimming velocities ranging from low to maximal. The overall F1-score of classification reached 0.96 with a temporal precision of 0.02 s. Lap times were directly computed from the classifier thanks to a high temporal precision and validated against a video gold standard. The mean absolute percentage error (MAPE) for this model against the video was 1.15%, 1%, and 4.07%, respectively, for starting lap times, middle lap times, and ending lap times. This model is a first step toward a powerful training assistant able to analyze swimmers with various levels of expertise in the context of in situ training monitoring.

Prediction of Kick Count in Triathletes during Freestyle Swimming Session Using Inertial Sensor Technology

Monitoring sports training performances with automatic, low cost, low power, and ergonomic solutions is a topic of increasing importance in the research of the last years. A parameter of particular interest, which has not been extensively dealt with in a state-of-the-art way, is the count of kicks during swimming training sessions. Coaches and athletes set the training sessions to optimize the kick count and swim stroke rate to acquire velocity and acceleration during swimming. In regard to race distances, counting kicks can influence the athlete’s performance. However, it is difficult to record the kick count without facing some issues about subjective interpretation. In this paper, a new method for kick count is proposed, based on only one triaxial accelerometer worn on the athlete’s ankle. The algorithm was validated on data recorded during freestyle training sessions. An accuracy of 97.5% with a sensitivity of 99.3% was achieved. The proposed method shows good linearity and a slope of 1.01. These results overcome other state-of-the-art methods, proving that this method is a good candidate for a reliable, embedded kick count.

Proposal of an Alpine Skiing Kinematic Analysis with the Aid of Miniaturized Monitoring Sensors, a Pilot Study

The recent growth and spread of smart sensor technologies make these connected devices suitable for diagnostic and monitoring in different fields. In particular, these sensors are useful in diagnostics for control of diseases or during rehabilitation. They are also extensively used in the monitoring field, both by non-expert and expert users, to monitor health status and progress during a sports activity. For athletes, these devices could be used to control and enhance their performance. This development has led to the realization of miniaturized sensors that are wearable during different sporting activities without interfering with the movements of the athlete. The use of these sensors, during training or racing, opens new frontiers for the understanding of motions and causes of injuries. This pilot study introduced a motion analysis system to monitor Alpine ski activities during training sessions. Through five inertial measurement units (IMUs), placed on five points of the athletes, it is possible to compute the angle of each joint and evaluate the ski run. Comparing the IMU data, firstly, with a video and then proposing them to an expert coach, it is possible to observe from the data the same mistakes visible in the camera. The aim of this work is to find a tool to support ski coaches during training sessions. Since the evaluation of athletes is now mainly developed with the support of video, we evaluate the use of IMUs to support the evaluation of the coach with more precise data.

The Use of Wearable Sensors for Preventing, Assessing, and Informing Recovery from Sport-Related Musculoskeletal Injuries: A Systematic Scoping Review

Wearable technologies are often indicated as tools that can enable the in-field collection of quantitative biomechanical data, unobtrusively, for extended periods of time, and with few spatial limitations. Despite many claims about their potential for impact in the area of injury prevention and management, there seems to be little attention to grounding this potential in biomechanical research linking quantities from wearables to musculoskeletal injuries, and to assessing the readiness of these biomechanical approaches for being implemented in real practice. We performed a systematic scoping review to characterise and critically analyse the state of the art of research using wearable technologies to study musculoskeletal injuries in sport from a biomechanical perspective. A total of 4952 articles were retrieved from the Web of Science, Scopus, and PubMed databases; 165 were included. Multiple study features-such as research design, scope, experimental settings, and applied context-were summarised and assessed. We also proposed an injury-research readiness classification tool to gauge the maturity of biomechanical approaches using wearables. Five main conclusions emerged from this review, which we used as a springboard to propose guidelines and good practices for future research and dissemination in the field.

Survey on Video-Based Biomechanics and Biometry Tools for Fracture and Injury Assessment in Sports

This work presents a survey literature review on biomechanics, specifically aimed at the study of existent biomechanical tools through video analysis, in order to identify opportunities for researchers in the field, and discuss future proposals and perspectives. Scientific literature (journal papers and conference proceedings) in the field of video-based biomechanics published after 2010 were selected and discussed. The most common application of the study of biomechanics using this technique is sports, where the most reported applications are american football, soccer, basketball, baseball, jumping, among others. These techniques have also been studied in a less proportion, in ergonomy, and injury prevention. From the revised literature, it is clear that biomechanics studies mainly focus on the analysis of angles, speed or acceleration, however, not many studies explore the dynamical forces in the joints. The development of video-based biomechanic tools for force analysis could provide methods for assessment and prediction of biomechanical force associated risks such as injuries and fractures. Therefore, it is convenient to start exploring this field. A few case studies are reported, where force estimation is performed via manual tracking in different scenarios. This demonstration is carried out using conventional manual tracking, however, the inclusion of similar methods in an automated manner could help in the development of intelligent healthcare, force prediction tools for athletes and/or elderly population. Future trends and challenges in this field are also discussed, where data availability and artificial intelligence models will be key to proposing new and more reliable methods for biomechanical analysis.

Translational Applications of Wearable Sensors in Education: Implementation and Efficacy

Background: Adding new approaches to teaching curriculums can be both expensive and complex to learn. The aim of this research was to gain insight into students’ literacy and confidence in learning sports science with new wearable technologies, specifically a novel program known as STEMfit. Methods: A three-phase design was carried out, with 36 students participating and exposed to wearable devices and associated software. This was to determine whether the technology hardware (phase one) and associated software (phase two) were used in a positive way that demonstrated user confidence. Results: Hardware included choosing a scalable wearable device that worked in conjunction with familiar and readily available software (Microsoft Excel) that extracted data through VBA coding. This allowed for students to experience and provide survey feedback on the usability and confidence gained when interacting with the STEMfit program. Outcomes indicated strong acceptance of the program, with high levels of motivation, resulting in a positive uptake of wearable technology as a teaching tool by students. The initial finding of this study offers an opportunity to further test the STEMfit program on other student cohorts as well as testing the scalability of the system into other year groups at the university level.

Integrated Timing of Stroking, Breathing, and Kicking in Front-Crawl Swimming: A Novel Stroke-by-Stroke Approach Using Wearable Inertial Sensors

Quantitative evaluation of synergic action among the different body segments is fundamental to swimming performance. The aim of the present study was to develop an easy-to-use tool for stroke-by-stroke evaluation of a swimmer’s integrated timing of stroking, kicking, and breathing. Twelve swimmers were evaluated during one trial of 100 m front-crawl swimming at self-selected speed. Five three-axial inertial sensors were mounted on the head, wrists, and ankles. Algorithms for the wrist entry into the water, the lower limb beat during the downward action, and the exit/entry of the face from/into the water were developed. Temporal events identified by video-based technique, using one sagittal moving camera, were assumed as the gold standard. The performance was evaluated in terms of the root-mean-square error, 90th percentile of absolute error, coefficient of variation, Bland–Altman plots, and correlation analysis. Results of all temporal events showed high agreement with the gold standard, confirmed by a root-mean-square error of less than 0.05 s for absolute temporal parameters and less than 0.7% for the percentages of the stroke cycle duration, and with correlation coefficients higher than 0.856. The protocol proposed was not only accurate and reliable, but also user-friendly and as unobtrusive as possible for the swimmer, allowing a stroke-by-stroke analysis during the training session.

The Effect of Paddle Stroke Variables Measured by Trainesense SmartPaddle® on the Velocity of the Kayak

(1) Background: This study aimed to compare key variables of paddle stroke measured by a commercial Trainesense SmartPaddle^® against the strain-gauge shaft and investigate how these variables are associated with the velocity of the boat among national-level canoe polo players. (2) Methods: This study involved 14 Finnish national-level canoe polo players. The measurement protocol consisted of three different paddling velocities, which were performed in indoor swimming pools. The velocity of the boat was calculated based on the performance time measured with the laser photocell gate. Canoe polo equipment was used in the study and a SmartPaddle sensor was attached to the paddle blade. A strain-gauge paddle shaft was used as a reference method to examine the validity of SmartPaddle. (3) Results: The stroke rate, force production time, mean and maximal force measured with the strain-gauge paddle shaft correlated strongly (r = 0.84–0.95, p < 0.01) with SmartPaddle. However, the SmartPaddle overestimated the maximum force compared to the strain-gauge shaft. Stroke rate (r = 0.86, p < 0.01), mean force (r = 0.79, p < 0.01), maximal force (r = 0.78, p < 0.01) and total absolute impulse (r = 0.70, p < 0.01) correlated positively and force production time negatively (r = −0.76, p < 0.01) with the velocity of the boat. (4) Conclusions: We conclude that the SmartPaddle provides promising information on stroke key variables when compared to the strain-gauge paddle shaft. The SmartPaddle is a new and interesting tool for biomechanical research and daily kayaking coaching in real open water conditions. However, more research and algorithm development are needed before the SmartPaddle can be used in everyday coaching sessions in kayaking.

Swimming Phase-Based Performance Evaluation Using a Single IMU in Main Swimming Techniques

Comprehensive monitoring of performance is essential for swimmers and swimming coaches to optimize the training. Regardless of the swimming technique, the swimmer passes various swimming phases from wall to wall, including a dive into the water or wall push-off, then glide and strokes preparation and finally, swimming up to the turn. The coach focuses on improving the performance of the swimmer in each of these phases. The purpose of this study was to assess the potential of using a sacrum-worn inertial measurement unit (IMU) for performance evaluation in each swimming phase (wall push-off, glide, stroke preparation and swimming) of elite swimmers in four main swimming techniques (i.e. front crawl, breaststroke, butterfly and backstroke). Nineteen swimmers were asked to wear a sacrum IMU and swim four one-way 25 m trials in each technique, attached to a tethered speedometer and filmed by cameras in the whole lap as reference systems. Based on the literature, several goal metrics were extracted from the instantaneous velocity (e.g. average velocity per stroke cycle) and displacement (e.g. time to reach 15 m from the wall) data from a tethered speedometer for the swimming phases, each one representing the goodness of swimmer’s performance. Following a novel approach, that starts from swimming bout detection and continues until detecting the swimming phases, the IMU kinematic variables in each swimming phase were extracted. The highly associated variables with the corresponding goal metrics were detected by LASSO (least absolute shrinkage and selection operator) variable selection and used for estimating the goal metrics with a linear regression model. The selected kinematic variables were relevant to the motion characteristics of each phase (e.g. selection of propulsion-related variables in wall push-off phase), providing more interpretability to the model. The estimation reached a determination coefficient (R²) value more than 0.75 and a relative RMSE less than 10% for most goal metrics in all swimming techniques. The results show that a single sacrum IMU can provide a wide range of performance-related swimming kinematic variables, useful for performance evaluation in four main swimming techniques.

Forces: A Motion Capture-Based Ergonomic Method for the Today's World

Approximately three of every five workers are affected by musculoskeletal disorders, especially in production environments. In this regard, workstation ergonomic evaluations are especially beneficial for conducting preventive actions. Nevertheless, today's context demonstrates that traditional ergonomic methods should lead to smart ergonomic methods. This document introduces the Forces ergonomic method, designed considering the possibilities of inertial motion capture technology and its applicability to evaluating actual workstations. This method calculates the joint risks for each posture and provides the total risk for the assessed workstation. In this calculation, Forces uses postural measurement and a kinetic estimation of all forces and torques that the joints support during movement. This paper details the method's fundamentals to achieve structural validity, demonstrating that all parts that compose it are logical and well-founded. This method aims to aid prevention technicians in focusing on what matters: making decisions to improve workers' health. Likewise, it aims to answer the current industry needs and reduce musculoskeletal disorders caused by repetitive tasks and lower the social, economic, and productivity losses that such disorders entail.

Validity, reliability and accuracy of inertial measurement units (IMUs) to measure angles: application in swimming

The first objective was to test the validity, reliability and accuracy of paired inertial measurement units (IMUs) to assess absolute angles relative to Vicon and OptiTrack systems. The potential impacts of slow vs. rapid and intermittent vs. continuous movements were tested during 2D laboratory analyses and 3D ecological context analysis. The second objective was to test the IMUs alone in an ecological activity (i.e., front crawl) that encompassed the previous independent variables to quantify inter-cyclic variability. Slow and intermittent motion ensured high to reasonable validity, reliability and accuracy. Rapid motion revealed an out-of-phase pattern for temporal reliability and lower validity, which was also visible in 3D. Also, spatial reliability and accuracy decreased in 3D, mainly due to discrepancies in local maximums, whereas temporal reliability remained in-phase. For the second objective, inter-cyclic variability did not exceed 12° based on root mean square error (RMSE). Therefore, IMUs should be considered valuable supplements to optoelectronic systems if users carefully position the sensors in rigid clusters and calibrate them to integrate potential offsets. Drift correction by spline interpolation or normalisation of the absolute data should also be considered as additional techniques that increase IMU performance in ecological contexts of performance.

Is machine learning and automatic classification of swimming data what unlocks the power of inertial measurement units in swimming?

Researchers have heralded the power of inertial sensors as a reliable swimmer-centric monitoring technology, however, regular uptake of this technology has not become common practice. Twenty-six elite swimmers participated in this study. An IMU (100Hz/500Hz) sensor was secured in the participant's third lumbar vertebrae. Features were extracted from swimming data using two techniques: a novel intrastroke cycle segmentation technique and conventional sliding window technique. Six supervised machine learning models were assessed on stroke prediction performance. Models trained using both feature extraction methods demonstrated high performance (≥ 0.99 weighted average precision, recall, F1-score, area under ROC curve and accuracy), low computational training times (< 3 seconds - bar XGB and when hyperparameters were tuned) and low computational prediction times (< 1 second). Significant differences were observed in weighted average stroke prediction F1-score (p = 0.0294) when using different feature extraction methods and model computational training time (p = 0.0007), and prediction time (p = 0.0026) when implementing hyperparameter tuning. Automatic swimming stroke classification offers benefits to observational coding and notational analysis, and opportunities for automated workload and performance monitoring in swimming. This stroke classification algorithm could be the key that unlocks the power of IMUs as a biofeedback tool in swimming.

Accelerometery vs. video-derived stroke parameters in high-level swimmers

Background

Swimming is a multifaceted sport with nuanced performance parameters that tend to vary according to the swimmer's stroke style. The extraction and analyses of swim parameters, such as lap time (LT), stroke length (SL), stroke rate (SR) and velocity are time-consuming. This may be eased and to some extent automated by the use of cost-effective tri-axial accelerometers.

Objectives

To determine the validity of tri-axial accelerometers across all four stroke styles, and to investigate kinematic differences in stroke styles using accelerometer-based data.

Methods

Twelve elite swimmers were recruited for the study. The group consisted of five male (age: 22.2 ± 2.6 years; height: 1.84 ± 0.08 m; weight: 76.2 ± 3.6 kg) and seven female (age: 20.7 ± 2.1 years; height: 1.68 ± 0.08 cm; weight: 62.0 ± 6.3 kg) swimmers.

Results

There was a small but significant bias for accelerometery data compared to video data across most parameters and stroke styles except for stroke length and stroke count (p > 0.05). However, accelerometery-derived SR, SL and velocity can be considered practically useful based on the training requirements of coaches. Parameters derived from video analysis compared to accelerometery were highly correlated (r > 0.91) and therefore consistent regardless of the method of analysis.

Conclusion

Although slight differences were present between the video and accelerometer data, these differences were not practically meaningful.

Involvement in Multiple Race Events Among International Para and Non-disabled Swimmers

International elite Para swimmers form a large portion of the overall multi-medalist winning population. For the highest performing Para swimmers, world class performances were achieved across different strokes. The aim of this study was to quantify the level of involvement across different events and to examine this in relation to the level of performance. The performances in swimming speed of the top 100 females and males for both Para- and non-disabled swimmers were collected in 11 race events between 2009 and 2019 (4,400 performances for 307 Para females and 365 Para males, 605 non-disabled females, and 715 non-disabled males). We tallied the number of events in which each swimmer was involved. Swimmers were grouped according to the total number of race events in which they participated. Then the association between involvement and level of performance was investigated. Para swimmers with impairment from classes seven to 14 were involved in a range of race events across different strokes. The most common combination for both Para and non-disabled athletes was over similarly distanced races of the same stroke (50 and 100 m freestyle). The more race events in which Para swimmers involved, the higher the level of performance that was achieved. This trend can partially be explained by the less concentrated competition pool for Para swimmers compared to able-bodied swimmers. Para swimmers with minimal and no physical impairment perform in multiple race events more often than able-bodied swimmers. Fewer Para swimmers at the international level and a less concentrated competition pool could explain these differences.

Instantaneous velocity estimation for the four swimming strokes using a 3-axis accelerometer: Validation on paralympic athletes

Inertial measurement unit systems (IMU) are increasingly used in sports. However, no algorithm evaluating the instantaneous swimming velocity has been validated on the four swimming strokes, or on dive start. Our objective is to develop and validate a new method to measure instantaneous swimming velocity on athletes presenting various musculoskeletal and neurological impairments and swimming one of the four swimming strokes. Seven Paralympic athletes were involved and performed a total of 18 trials of 50 m in real conditions. All trials were recorded with a 3-axis accelerometer, a tethered device and a video camera. The instantaneous velocity was computed by drift-free integration of the forward acceleration. For all trials, Bland-Altman analyses showed a bias of 0.03-0.06 m.s^-1 with the tethered device, with 95% LOA lower than 0.31-0.80 m.s^-1, RMSE was 0.14-0.39 m.s^-1, and ICC amounted 0.494-0.941. No significant difference was found for the mean 50 m velocities between the accelerometer, the tethered device and the video camera. The results obtained for freestyle were comparable to those obtained with GPS in outdoor pools. Those obtained for backstroke, breaststroke and butterfly were very encouraging. Our method is simple, reliable, advantageous compared to tethered devices, and could be easily used in the field by coaches.

A Novel Macro-Micro Approach for Swimming Analysis in Main Swimming Techniques Using IMU Sensors

Inertial measurement units (IMU) are proven as efficient tools for swimming analysis by overcoming the limits of video-based systems application in aquatic environments. However, coaches still believe in the lack of a reliable and easy-to-use analysis system for swimming. To provide a broad view of swimmers' performance, this paper describes a new macro-micro analysis approach, comprehensive enough to cover a full training session, regardless of the swimming technique. Seventeen national level swimmers (5 females, 12 males, 19.6 ± 2.1 yrs) were equipped with six IMUs and asked to swim 4 × 50 m trials in each swimming technique (i.e., frontcrawl, breaststroke, butterfly, and backstroke) in a 25 m pool, in front of five 2-D cameras (four under water and one over water) for validation. The proposed approach detects swimming bouts, laps, and swimming technique in macro level and swimming phases in micro level on all sensor locations for comparison. Swimming phases are the phases swimmers pass from wall to wall (wall push-off, glide, strokes preparation, swimming, and turn) and micro analysis detects the beginning of each phase. For macro analysis, an overall accuracy range of 0.83–0.98, 0.80–1.00, and 0.83–0.99 were achieved, respectively, for swimming bouts detection, laps detection and swimming technique identification on selected sensor locations, the highest being achieved with sacrum. For micro analysis, we obtained the lowest error mean and standard deviation on sacrum for the beginning of wall-push off, glide and turn (−20 ± 89 ms, 4 ± 100 ms, 23 ± 97 ms, respectively), on shank for the beginning of strokes preparation (0 ± 88 ms) and on wrist for the beginning of swimming (−42 ± 72 ms). Comparing the swimming techniques, sacrum sensor achieves the smallest range of error mean and standard deviation during micro analysis. By using the same macro-micro approach across different swimming techniques, this study shows its efficiency to detect the main events and phases of a training session. Moreover, comparing the results of both macro and micro analyses, sacrum has achieved relatively higher amounts of accuracy and lower mean and standard deviation of error in all swimming techniques.

Arm-Stroke Descriptor Variability during 200-m Front Crawl Swimming

The present study aimed to explore the variability of the arm-stroke temporal descriptors between and within laps during middle-distance swimming event using IMMUs. Eight male swimmers performed a 200-m maximum front-crawl in which the inter-lap and intra-lap variability of velocity, stroke rate, stroke-phases duration and arm-coordination index were measured through five units of IMMU. An algorithm computes the 3D coordinates of the wrist by means the IMMU orientation and the kinematic chain of upper arm biomechanical model, and it recognizes the start events of the four arm-stroke phases. Velocity and stroke rate had a mean value of 1.47 ± 0.10 m·s⁻¹ and 32.94 ± 4.84 cycles·min⁻¹, respectively, and a significant decrease along the 200-m (p < 0.001; η² = 0.80 and 0.47). The end of each lap showed significantly lower stroke rate compared to the start and the middle segment (p < 0.05; η² = 0.55). No other significant inter-lap and intra-lap differences were detected. The two main findings are: (i) IMMUs technology can be an effective solution to continuously monitor the temporal descriptors during the swimming trial; (ii) swimmers are able to keep stable their temporal technique descriptors in a middle-distance event, despite the decrease of velocity and stroke rate.

Optimizing Sensor Deployment for Multi-Sensor-Based HAR System with Improved Glowworm Swarm Optimization Algorithm

Human activity recognition (HAR) technology that analyzes and fuses the data acquired from various homogeneous or heterogeneous sensor sources has motivated the development of enormous human-centered applications such as healthcare, fitness, ambient assisted living and rehabilitation. The concurrent use of multiple sensor sources for HAR is a good choice because the plethora of user information provided by the various sensor sources may be useful. However, a multi-sensor system with too many sensors will bring large power consumption and some sensor sources may bring little improvements to the performance. Therefore, the multi-sensor deployment research that can gain a tradeoff among computational complexity and performance is imperative. In this paper, we propose a multi-sensor-based HAR system whose sensor deployment can be optimized by selective ensemble approaches. With respect to optimization of the sensor deployment, an improved binary glowworm swarm optimization (IBGSO) algorithm is proposed and the sensor sources that have a significant effect on the performance of HAR are selected. Furthermore, the ensemble learning system based on optimized sensor deployment is constructed for HAR. Experimental results on two datasets show that the proposed IBGSO-based multi-sensor deployment approach can select a smaller number of sensor sources while achieving better performance than the ensemble of all sensors and other optimization-based selective ensemble approaches.

The current use of wearable sensors to enhance safety and performance in breath-hold diving: A systematic review

INTRODUCTION

Measuring physiological parameters at depth is an emergent challenge for athletic training, diver's safety and biomedical research. Recent advances in wearable sensor technology made this challenge affordable; however, its impact on breath-hold diving has never been comprehensively discussed.

METHODS

We performed a systematic review of the literature in order to assess what types of sensors are available or suitable for human breath-hold diving, within the two-fold perspective of safety and athletic performance.

RESULTS

In the 52 studies identified, sensed physiological variables were: electrocardiogram, body temperature, blood pressure, peripheral oxygen saturation, interstitial glucose concentration, impedance cardiography, heart rate, body segment inertia and orientation.

CONCLUSIONS

Limits and potential of each technology are separately reviewed. Inertial sensor technology and transmission pulse oximetry could produce the greatest impact on breath-hold diving performances in the future.

Gait Analysis in a Box: A System Based on Magnetometer-Free IMUs or Clusters of Optical Markers with Automatic Event Detection

Gait analysis based on full-body motion capture technology (MoCap) can be used in rehabilitation to aid in decision making during treatments or therapies. In order to promote the use of MoCap gait analysis based on inertial measurement units (IMUs) or optical technology, it is necessary to overcome certain limitations, such as the need for magnetically controlled environments, which affect IMU systems, or the need for additional instrumentation to detect gait events, which affects IMUs and optical systems. We present a MoCap gait analysis system called Move Human Sensors (MH), which incorporates proposals to overcome both limitations and can be configured via magnetometer-free IMUs (MH-IMU) or clusters of optical markers (MH-OPT). Using a test-retest reliability experiment with thirty-three healthy subjects (20 men and 13 women, 21.7 ± 2.9 years), we determined the reproducibility of both configurations. The assessment confirmed that the proposals performed adequately and allowed us to establish usage considerations. This study aims to enhance gait analysis in daily clinical practice.

Assessing the Effects of Kata and Kumite Techniques on Physical Performance in Elite Karatekas

This study aimed at assessing physical performance of elite karatekas and non-karatekas. More specifically, effects of kumite and kata technique on joint mobility, body stability, and jumping ability were assessed by enrolling twenty-four karatekas and by comparing the results with 18 non-karatekas healthy subjects. Sensor system was composed by a single inertial sensor and optical bars. Karatekas are generally characterized by better motor performance with respect nonkaratekas, considering all the examined factors, i.e., mobility, stability, and jumping. In addition, the two techniques lead to a differentiation in joint mobility; in particular, kumite athletes are characterized by a greater shoulder extension and, in general, by a greater value of preferred velocity to perform joint movements. Conversely, kata athletes are characterized by a greater mobility of the ankle joint. By focusing on jumping skills, kata technique leads to an increase of the concentric phase when performing squat jump. Finally, kata athletes showed better stability inclosed eyes condition. The outcomes reported here can be useful for optimizing coaching programs for both beginners and karatekas based on the specific selected technique.

Inertial Sensor-Based Motion Tracking in Football with Movement Intensity Quantification

Inertial sensor-based measurements of lower body kinematics in football players may improve physical load estimates during training sessions and matches. However, the validity of inertial-based motion analysis systems is specific to both the type of movement and the intensity at which movements are executed. Importantly, such a system should be relatively simple, so it can easily be used in daily practice. This paper introduces an easy-to-use inertial-based motion analysis system and evaluates its validity using an optoelectronic motion analysis system as a gold standard. The system was validated in 11 football players for six different football specific movements that were executed at low, medium, and maximal intensity. Across all movements and intensities, the root mean square differences (means ± SD) for knee and hip flexion/extension angles were 5.3° ± 3.4° and 8.0° ± 3.5°, respectively, illustrating good validity with the gold standard. In addition, mean absolute flexion/extension angular velocities significantly differed between the three movement intensities. These results show the potential to use the inertial based motion analysis system in football practice to obtain lower body kinematics and to quantify movement intensity, which both may improve currently used physical load estimates of the players.

A Pilot Study on Continuous Breaststroke Phase Recognition with Fast Training Based on Lower-Limb Inertial Signals

In this study, we proposed a continuous stroke phase recognition method with lower-limb inertial signals. The aim of the method was to decrease the time needed and to relieve the burdensome manual configurations in the tasks of human underwater motion recognition. The method automatically segmented the data of a period of time into stroke cycles and three sub-phases (propulsion, glide and recovery). K-nearest neighbor algorithm (k-NN) was used as the classifier to train the segmented data and classify the new data on each sample interval. To validate the proposed recognition method, three elite swimmers were recruited. We also designed an wearable sensing system for human underwater motion sensing with inertial measurement units (IMUs). With only data of 5 stroke cycles for training, the recognizer produced accurate recognition results. The average precision across the phases and the subjects was 93.7% and the average recall was 92.6%. We also investigated the time difference of the key stroke events (stroke phase transitions) between the recognized decisions and the reference ones. The average time difference was 66.2 ms, which accounted for the 4.2% of a single stroke phase. The results of the pilot study proved the feasibility of the new method for human aquatic locomotion assistance tasks. Future efforts will be paid in this new direction for more promising results.

The Use of Wearable Sensors in Human Movement Analysis in Non-Swimming Aquatic Activities: A Systematic Review

The use of smart technology, specifically inertial sensors (accelerometers, gyroscopes, and magnetometers), to analyze swimming kinematics is being reported in the literature. However, little is known about the usage/application of such sensors in other human aquatic exercises. As the sensors are getting smaller, less expensive, and simple to deal with (regarding data acquisition), one might consider that its application to a broader range of exercises should be a reality. The aim of this systematic review was to update the state of the art about the framework related to the use of sensors assessing human movement in an aquatic environment, besides swimming. The following databases were used: IEEE Xplore, Pubmed, Science Direct, Scopus, and Web of Science. Five articles published in indexed journals, aiming to assess human exercises/movements in the aquatic environment were reviewed. The data from the five articles was categorized and summarized based on the aim, purpose, participants, sensor's specifications, body area and variables analyzed, and data analysis and statistics. The analyzed studies aimed to compare the movement/exercise kinematics between environments (i.e., dry land versus aquatic), and in some cases compared healthy to pathological participants. The use of sensors in a rehabilitation/hydrotherapy perspective may provide major advantages for therapists.

Integrating a gait analysis test in hospital rehabilitation: A service design approach

BACKGROUND

Gait analysis with motion capture (MoCap) during rehabilitation can provide objective information to facilitate treatment decision making. However, designing a test to be integrated into healthcare services requires considering multiple design factors. The difficulty of integrating a 'micro-service' (gait test) within a 'macro-service' (healthcare service) has received little attention in the gait analysis literature. It is a challenge that goes beyond the gait analysis case study because service design methods commonly focus on the entire service design (macro-level).

OBJECTIVE

This study aims to extract design considerations and generate guidelines to integrate MoCap technology for gait analysis in the hospital rehabilitation setting. Specifically, the aim is to design a gait test to assess the response of the applied treatments through pre- and post-measurement sessions.

METHODS

We focused on patients with spasticity who received botulinum toxin treatment. A qualitative research design was used to investigate the integration of a gait analysis system based on inertial measurement units in a rehabilitation service at a reference hospital. The methodological approach was based on contrasted methodologies from the service design field, which materialise through observation techniques (during system use), semi-structured interviews, and workshops with healthcare professionals (13 patients, 10 'proxies', and 6 doctors).

RESULTS

The analysis resulted in six themes: (1) patients' understanding, (2) guiding the gait tests, (3) which professionals guide the gait tests, (4) gait test reports, (5) requesting gait tests (doctors and test guide communication), and the (6) conceptual design of the service with the gait test.

CONCLUSIONS

The extracted design considerations and guidelines increase the applicability and usefulness of the gait analysis technology, improving the link between technologists and healthcare professionals. The proposed methodological approach can also be useful for service design teams that deal with the integration of one service into another.

Selective Ensemble Based on Extreme Learning Machine for Sensor-Based Human Activity Recognition

Sensor-based human activity recognition (HAR) has attracted interest both in academic and applied fields, and can be utilized in health-related areas, fitness, sports training, etc. With a view to improving the performance of sensor-based HAR and optimizing the generalizability and diversity of the base classifier of the ensemble system, a novel HAR approach (pairwise diversity measure and glowworm swarm optimization-based selective ensemble learning, DMGSOSEN) that utilizes ensemble learning with differentiated extreme learning machines (ELMs) is proposed in this paper. Firstly, the bootstrap sampling method is utilized to independently train multiple base ELMs which make up the initial base classifier pool. Secondly, the initial pool is pre-pruned by calculating the pairwise diversity measure of each base ELM, which can eliminate similar base ELMs and enhance the performance of HAR system by balancing diversity and accuracy. Then, glowworm swarm optimization (GSO) is utilized to search for the optimal sub-ensemble from the base ELMs after pre-pruning. Finally, majority voting is utilized to combine the results of the selected base ELMs. For the evaluation of our proposed method, we collected a dataset from different locations on the body, including chest, waist, left wrist, left ankle and right arm. The experimental results show that, compared with traditional ensemble algorithms such as Bagging, Adaboost, and other state-of-the-art pruning algorithms, the proposed approach is able to achieve better performance (96.7% accuracy and F1 from wrist) with fewer base classifiers.

Wearable sensors for monitoring the internal and external workload of the athlete

The convergence of semiconductor technology, physiology, and predictive health analytics from wearable devices has advanced its clinical and translational utility for sports. The detection and subsequent application of metrics pertinent to and indicative of the physical performance, physiological status, biochemical composition, and mental alertness of the athlete has been shown to reduce the risk of injuries and improve performance and has enabled the development of athlete-centered protocols and treatment plans by team physicians and trainers. Our discussions in this review include commercially available devices, as well as those described in scientific literature to provide an understanding of wearable sensors for sports medicine. The primary objective of this paper is to provide a comprehensive review of the applications of wearable technology for assessing the biomechanical and physiological parameters of the athlete. A secondary objective of this paper is to identify collaborative research opportunities among academic research groups, sports medicine health clinics, and sports team performance programs to further the utility of this technology to assist in the return-to-play for athletes across various sporting domains. A companion paper discusses the use of wearables to monitor the biochemical profile and mental acuity of the athlete.

Anaerobic capacity assessment in elite swimmers through inertial sensors

OBJECTIVE

The present study aimed to assess if changes in speed and stroke parameters, as measured by an inertial sensor during a maximal effort swimming test, could provide an effective detection of anaerobic capacity in elite swimmers.

APPROACH

Fourteen elite swimmers performed a 75 m maximal swimming test. Changes in speed and stroke parameters, estimated by a body-worn inertial sensor, were analysed to provide insight into stroke mechanics during swimming. Their relationships with the output of the Wingate Anaerobic Test were analysed. Best times in competition were also considered to assess swimmer's performance.

MAIN RESULTS

Mean power measured using the Wingate cycle ergometer test highly correlated with mean speed attained by the swimmers during the proposed 75 m swimming test (R range: .700-.809, p   <  .05). Mean power in the Wingate Anaerobic Test and mean speed in the 75 m swimming test highly correlated with best times attained by the swimmers (R range: .736-.855, p   <  .01; R range: .659-.952, p   <  .05, for Wingate and 75 m swimming test, respectively). Moreover, stroke variables were investigated: in this regard, a significant decrease in stroke rate and swimming speed and a significant increase in stroke length were observed between the first and the third lap (p   <  .01).

SIGNIFICANCE

The present in-water free swimming test provided insight into specific physiological/mechanical aspects of elite swimmers. The correlation of the swimming and the Wingate tests with swimmer's performance in competition confirms that they both reflect the skills and anaerobic qualities a swimmer uses in a race. The wearable inertial sensor could represent a feasible solution to evaluate stroke parameters, allowing a timely follow-up of variations in swimming biomechanics along the course of the test and the identification of differences in biomechanical strategy between swimmers. This analysis is of great interest for swimmers and coaches to characterise swimmer's technique weakness and strength, and to plan individual race pacing strategy.

Wearable Sensor-Based Human Activity Recognition via Two-Layer Diversity-Enhanced Multiclassifier Recognition Method

Sensor-based human activity recognition can benefit a variety of applications such as health care, fitness, smart homes, rehabilitation training, and so forth. In this paper, we propose a novel two-layer diversity-enhanced multiclassifier recognition method for single wearable accelerometer-based human activity recognition, which contains data-based and classifier-based diversity enhancement. Firstly, we introduce the kernel Fisher discriminant analysis (KFDA) technique to spatially transform the training samples and enhance the discrimination between activities. In addition, bootstrap resampling is utilized to increase the diversities of the dataset for training the base classifiers in the multiclassifier system. Secondly, a combined diversity measure for selecting the base classifiers with excellent performance and large diversity is proposed to optimize the performance of the multiclassifier system. Lastly, majority voting is utilized to combine the preferred base classifiers. Experiments showed that the data-based diversity enhancement can improve the discriminance of different activity samples and promote the generation of base classifiers with different structures and performances. Compared with random selection and traditional ensemble methods, including Bagging and Adaboost, the proposed method achieved 92.3% accuracy and 90.7% recall, which demonstrates better performance in activity recognition.

Measurement of Upper Limb Range of Motion Using Wearable Sensors: A Systematic Review

Background

Wearable sensors are portable measurement tools that are becoming increasingly popular for the measurement of joint angle in the upper limb. With many brands emerging on the market, each with variations in hardware and protocols, evidence to inform selection and application is needed. Therefore, the objectives of this review were related to the use of wearable sensors to calculate upper limb joint angle. We aimed to describe (i) the characteristics of commercial and custom wearable sensors, (ii) the populations for whom researchers have adopted wearable sensors, and (iii) their established psychometric properties.

Methods

A systematic review of literature was undertaken using the following data bases: MEDLINE, EMBASE, CINAHL, Web of Science, SPORTDiscus, IEEE, and Scopus. Studies were eligible if they met the following criteria: (i) involved humans and/or robotic devices, (ii) involved the application or simulation of wearable sensors on the upper limb, and (iii) calculated a joint angle.

Results

Of 2191 records identified, 66 met the inclusion criteria. Eight studies compared wearable sensors to a robotic device and 22 studies compared to a motion analysis system. Commercial (n = 13) and custom (n = 7) wearable sensors were identified, each with variations in placement, calibration methods, and fusion algorithms, which were demonstrated to influence accuracy.

Conclusion

Wearable sensors have potential as viable instruments for measurement of joint angle in the upper limb during active movement. Currently, customised application (i.e. calibration and angle calculation methods) is required to achieve sufficient accuracy (error <  5°). Additional research and standardisation is required to guide clinical application.

Trial Registration

This systematic review was registered with PROSPERO (CRD42017059935).