scholarly journals Towards Machine Learning-Based Detection of Running-Induced Fatigue in Real-World Scenarios: Evaluation of IMU Sensor Configurations to Reduce Intrusiveness

Sensors ◽  
2021 ◽  
Vol 21 (10) ◽  
pp. 3451
Author(s):  
Luca Marotta ◽  
Jaap H. Buurke ◽  
Bert-Jan F. van Beijnum ◽  
Jasper Reenalda
Keyword(s):  
Machine Learning ◽  
Perceived Exertion ◽  
Inertial Sensors ◽  
Moving Average ◽  
Progressive Increase ◽  
Classification Algorithm ◽  
Rating Of Perceived Exertion ◽  
Left Tibia ◽  
Machine Learning Classification ◽  
Increased Risk

Physical fatigue is a recurrent problem in running that negatively affects performance and leads to an increased risk of being injured. Identification and management of fatigue helps reducing such negative effects, but is presently commonly based on subjective fatigue measurements. Inertial sensors can record movement data continuously, allowing recording for long durations and extensive amounts of data. Here we aimed to assess if inertial measurement units (IMUs) can be used to distinguish between fatigue levels during an outdoor run with a machine learning classification algorithm trained on IMU-derived biomechanical features, and what is the optimal configuration to do so. Eight runners ran 13 laps of 400 m on an athletic track at a constant speed with 8 IMUs attached to their body (feet, tibias, thighs, pelvis, and sternum). Three segments were extracted from the run: laps 2–4 (no fatigue condition, Rating of Perceived Exertion (RPE) = 6.0 ± 0.0); laps 8–10 (mild fatigue condition, RPE = 11.7 ± 2.0); laps 11–13 (heavy fatigue condition, RPE = 14.2 ± 3.0), run directly after a fatiguing protocol (progressive increase of speed until RPE ≥ 16) that followed lap 10. A random forest classification algorithm was trained with selected features from the 400 m moving average of the IMU-derived accelerations, angular velocities, and joint angles. A leave-one-subject-out cross validation was performed to assess the optimal combination of IMU locations to detect fatigue and selected sensor configurations were considered. The left tibia was the most recurrent sensor location, resulting in accuracies ranging between 0.761 (single left tibia location) and 0.905 (all IMU locations). These findings contribute toward a balanced choice between higher accuracy and lower intrusiveness in the development of IMU-based fatigue detection devices in running.

scholarly journals Protection Against Spikes in Workload With Aerobic Fitness and Playing Experience: The Role of the Acute:Chronic Workload Ratio on Injury Risk in Elite Gaelic Football

2017 ◽  
Vol 12 (3) ◽  
pp. 393-401 ◽  
Author(s):  
Shane Malone ◽  
Mark Roe ◽  
Dominic A. Doran ◽  
Tim J. Gabbett ◽  
Kieran D. Collins
Keyword(s):  
Aerobic Fitness ◽  
Injury Risk ◽  
Perceived Exertion ◽  
Reference Group ◽  
Time Trial ◽  
Rating Of Perceived Exertion ◽  
Football Players ◽  
Risk Of Injury ◽  
Increased Risk ◽  
Injury Protection

Purpose:To examine the association between combined session rating of perceived exertion (RPE) workload measures and injury risk in elite Gaelic footballers.Methods:Thirty-seven elite Gaelic footballers (mean ± SD age 24.2 ± 2.9 y) from 1 elite squad were involved in a single-season study. Weekly workload (session RPE multiplied by duration) and all time-loss injuries (including subsequent-wk injuries) were recorded during the period. Rolling weekly sums and wk-to-wk changes in workload were measured, enabling the calculation of the acute:chronic workload ratio by dividing acute workload (ie, 1-weekly workload) by chronic workload (ie, rolling-average 4-weekly workload). Workload measures were then modeled against data for all injuries sustained using a logistic-regression model. Odds ratios (ORs) were reported against a reference group.Results:High 1-weekly workloads (≥2770 arbitrary units [AU], OR = 1.63–6.75) were associated with significantly higher risk of injury than in a low-training-load reference group (<1250 AU). When exposed to spikes in workload (acute:chronic workload ratio >1.5), players with 1 y experience had a higher risk of injury (OR = 2.22) and players with 2–3 (OR = 0.20) and 4–6 y (OR = 0.24) of experience had a lower risk of injury. Players with poorer aerobic fitness (estimated from a 1-km time trial) had a higher injury risk than those with higher aerobic fitness (OR = 1.50–2.50). An acute:chronic workload ratio of (≥2.0) demonstrated the greatest risk of injury.Conclusions:These findings highlight an increased risk of injury for elite Gaelic football players with high (>2.0) acute:chronic workload ratios and high weekly workloads. A high aerobic capacity and playing experience appears to offer injury protection against rapid changes in workload and high acute:chronic workload ratios. Moderate workloads, coupled with moderate to high changes in the acute:chronic workload ratio, appear to be protective for Gaelic football players.

Various Workload Models and the Preseason Are Associated With Injuries in Professional Female Cyclists

2021 ◽  
pp. 1-6
Author(s):  
Teun van Erp ◽  
Taco van der Hoorn ◽  
Marco J.M. Hoozemans ◽  
Carl Foster ◽  
Jos J. de Koning
Keyword(s):  
Generalized Estimating Equations ◽  
Perceived Exertion ◽  
Estimating Equations ◽  
Rating Of Perceived Exertion ◽  
Moving Averages ◽  
The Third ◽  
Increased Risk ◽  
The Difference ◽  
Workload Models ◽  
Exponentially Weighted

Purpose: To determine if workload and seasonal periods (preseason vs in season) are associated with the incidence of injuries and illnesses in female professional cyclists. Methods: Session rating of perceived exertion was used to quantify internal workload and was collected from 15 professional female cyclists, from 33 athlete seasons. One week (acute) workload, 4 weeks (chronic) workload, and 3 acute:chronic workload models were analyzed. Two workload models are based on moving averages of the ratios, the acute:chronic workload ratio (ACWR), and the ACWR uncoupled (ACWRuncoup). The difference between both is the chronic load; in ACWR, the acute load is part of the chronic load, and in ACWRuncoup, the acute and chronic load are uncoupled. The third workload model is based on exponentially weighted moving averages of the ratios. In addition, the athlete season is divided into the preseason and in season. Results: Generalized estimating equations analysis was used to assess the associations between the workload ratios and the occurrence of injuries and illnesses. High values of acute workload (P = .048), ACWR (P = .02), ACWRuncoup (P = .02), exponentially weighted moving averages of the ratios (P = .01), and the in season (P = .0001) are significantly associated with the occurrence of injury. No significant associations were found between the workload models, the seasonal periods, and the occurrence of illnesses. Conclusions: These findings suggest the importance of monitoring workload and workload ratios in female professional cyclists to lower the risk of injuries and therefore improve their performances. Furthermore, these results indicate that, in the preseason, additional stressors occur, which could lead to an increased risk of injuries.

scholarly journals Changes in Markers of Oxidative Stress and α-Amylase in Saliva of Children Associated with a Tennis Competition

2020 ◽  
Vol 17 (17) ◽  
pp. 6269 ◽  
Author(s):  
José María Giménez-Egido ◽  
Raquel Hernández-García ◽  
Damián Escribano ◽  
Silvia Martínez-Subiela ◽  
Gema Torres-Luque ◽  
...  
Keyword(s):  
Oxidative Stress ◽  
Antioxidant Capacity ◽  
Psychological Stress ◽  
Antioxidant Status ◽  
Perceived Exertion ◽  
Progressive Increase ◽  
Rating Of Perceived Exertion ◽  
Trolox Equivalent Antioxidant Capacity ◽  
Reducing Ability ◽  
Trolox Equivalent

The purpose of this paper was to analyze the changes caused by a one-day tennis tournament in biomarkers of oxidative stress and α-amylase in saliva in children. The sample was 20 male active children with the following characteristics: (a) age of players = 9.46 ± 0.66 years; (b) weight = 34.8 ± 6.5 kg; (c) height = 136.0 ± 7.9 cm; (d) mean weekly training tennis = 2.9 ± 1.0 h. The tennis competition ran for one day, with four matches for each player. Data were taken from the average duration per match and the rating of perceived exertion (RPE). Four biomarkers of antioxidant status: uric acid (AU), Trolox equivalent antioxidant capacity (TEAC), ferric reducing ability of saliva (FRAS, cupric reducing antioxidant capacity (CUPRAC) and salivary alpha-amylase (sAA) as a biomarker of psychological stress were measured in saliva. The time points were baseline (at home before the tournament), pre-competition (immediately before the first match) and post-match (after each match) measurements. The four biomarkers of antioxidant status showed a similar dynamic with lower values at baseline and a progressive increase during the four matches. Overall one-day tennis competition in children showed a tendency to increase antioxidant biomarkers in saliva. In addition, there was an increase in pre-competition sAA possibly associated with psychological stress. Further studies about the possible physiological implications of these findings should be performed in the future.

Machine Learning Classification of Head Impact Sensor Data

2019 ◽  
Author(s):  
Tyler F. Rooks ◽  
Andrea S. Dargie ◽  
Valeta Carol Chancey
Keyword(s):  
Machine Learning ◽  
Decision Tree ◽  
External Validation ◽  
Classification Algorithm ◽  
Sensor Data ◽  
Environmental Sensors ◽  
Head Acceleration ◽  
Machine Learning Classification ◽  
Environmental Sensor ◽  
Validation Set

Abstract A shortcoming of using environmental sensors for the surveillance of potentially concussive events is substantial uncertainty regarding whether the event was caused by head acceleration (“head impacts”) or sensor motion (with no head acceleration). The goal of the present study is to develop a machine learning model to classify environmental sensor data obtained in the field and evaluate the performance of the model against the performance of the proprietary classification algorithm used by the environmental sensor. Data were collected from Soldiers attending sparring sessions conducted under a U.S. Army Combatives School course. Data from one sparring session were used to train a decision tree classification algorithm to identify good and bad signals. Data from the remaining sparring sessions were kept as an external validation set. The performance of the proprietary algorithm used by the sensor was also compared to the trained algorithm performance. The trained decision tree was able to correctly classify 95% of events for internal cross-validation and 88% of events for the external validation set. Comparatively, the proprietary algorithm was only able to correctly classify 61% of the events. In general, the trained algorithm was better able to predict when a signal was good or bad compared to the proprietary algorithm. The present study shows it is possible to train a decision tree algorithm using environmental sensor data collected in the field.

A synthetic undergarment increases physiological strain

2019 ◽  
Vol 28 (4) ◽  
pp. 275 ◽  
Author(s):  
Matthew C. Dorton ◽  
Brent C. Ruby ◽  
Charles L. Dumke
Keyword(s):  
Blood Flow ◽  
Skin Blood Flow ◽  
Perceived Exertion ◽  
Sweat Rate ◽  
Rest Period ◽  
Time Trial ◽  
Rating Of Perceived Exertion ◽  
Physiological Strain ◽  
Physiological Factors ◽  
Increased Risk

Our aim was to examine the effect of a synthetic material undergarment on heat stress during exercise in a hot environment. Ten active males completed two trials of intermittent (50min walking, 10min sitting) treadmill walking over 3h in 35°C and 30% relative humidity. Subjects wore wildland firefighter flame-resistant meta-aramid blend pants and shirt with either a 100% cotton (C) or flame-retardant modacrylic undergarment (S), while carrying a 16-kg pack, helmet and leather gloves. Exercise was followed by a 30-min rest period without pack, helmet, gloves, and outerwear shirt. Rectal temperature and physiological strain were greater in S than C (P=0.04). No significant differences were found for heart rate, rating of perceived exertion, energy expenditure or skin temperature between C and S. Skin blood flow increased significantly in S following the second hour of exercise, resulting in a time×trial interaction (P=0.001). No significant differences for skin blood flow were found post exercise. Sweat rate and percent dehydration were not different between C and S. These data indicate that, of the two undergarments investigated, the synthetic undergarment negatively affected physiological factors that have been shown to indicate an increased risk of heat-related injuries.

scholarly journals Fake Job Detection and Analysis Using Machine Learning and Deep Learning Algorithms

2021 ◽  
Vol 11 (2) ◽  
pp. 642-650
Author(s):  
C.S. Anita ◽  
P. Nagarajan ◽  
G. Aditya Sairam ◽  
P. Ganesh ◽  
G. Deepakkumar
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Data Cleaning ◽  
Personal Information ◽  
Learning Algorithms ◽  
Classification Algorithm ◽  
Paper Machine ◽  
Machine Learning Classification ◽  
Accuracy And Precision ◽  
Processing Step

With the pandemic situation, there is a strong rise in the number of online jobs posted on the internet in various job portals. But some of the jobs being posted online are actually fake jobs which lead to a theft of personal information and vital information. Thus, these fake jobs can be precisely detected and classified from a pool of job posts of both fake and real jobs by using advanced deep learning as well as machine learning classification algorithms. In this paper, machine learning and deep learning algorithms are used so as to detect fake jobs and to differentiate them from real jobs. The data analysis part and data cleaning part are also proposed in this paper, so that the classification algorithm applied is highly precise and accurate. It has to be noted that the data cleaning step is a very important step in machine learning project because it actually determines the accuracy of the machine learning as well as deep learning algorithms. Hence a great importance is emphasized on data cleaning and pre-processing step in this paper. The classification and detection of fake jobs can be done with high accuracy and high precision. Hence the machine learning and deep learning algorithms have to be applied on cleaned and pre-processed data in order to achieve a better accuracy. Further, deep learning neural networks are used so as to achieve higher accuracy. Finally all these classification models are compared with each other to find the classification algorithm with highest accuracy and precision.

Sign in / Sign up

Export Citation Format

Share Document