Predicting Injury and Illness with Machine Learning in Elite Youth Soccer: A Comprehensive Monitoring Approach over 3 Months.

The search for monitoring tools that provide early indication of injury and illness could contribute to better player protection. The aim of the present study was to i) determine the feasibility of and adherence to our monitoring approach, and ii) identify variables associated with up-coming illness and injury. We incorporated a comprehensive set of monitoring tools consisting of external load and physical fitness data, questionnaires, blood, neuromuscular-, hamstring, hip abductor and hip adductor performance tests performed over a three-month period in elite under-18 academy soccer players. Twenty-five players (age: 16.6 ± 0.9 years, height: 178 ± 7 cm, weight: 74 ± 7 kg, VO2max: 59 ± 4 ml/min/kg) took part in the study. In addition to evaluating adherence to the monitoring approach, data were analyzed using a linear support vector machine (SVM) to predict illness and injuries. The approach was feasible, with no injuries or dropouts due to the monitoring process. Questionnaire adherence was high at the beginning and decreased steadily towards the end of the study. An SVM resulted in the best classification results for three classification tasks, i.e., illness prediction, illness determination and injury prediction. For injury prediction, one of four injuries present in the test data set was detected, with 96.3% of all data points (i.e., injuries and non-injuries) correctly detected. For both illness prediction and determination, there was only one illness in the test data set that was detected by the linear SVM. However, the model showed low precision for injury and illness prediction with a considerable number of false-positives. The results demonstrate the feasibility of a holistic monitoring approach with the possibility of predicting illness and injury. Additional data points are needed to improve the prediction models. In practical application, this may lead to overcautious recommendations on when players should be protected from injury and illness.

Assessment of Exercise Intensity for Uphill Walking in Healthy Adults Performed Indoors and Outdoors.

BACKGROUND: Borg's rating of perceived exertion (BRPE) scale is a simple, but subjective tool to grade physical strain during exercise. As a result, it is widely used for the prescription of exercise intensity, especially for cardiovascular disease prevention. The purpose of this study was to assess and compare relationships between BRPE and physiological measures of exercise intensity during uphill walking indoors and outdoors. METHODS: 134 healthy participants [median age: 56 years (IQR 52-63)] completed a maximal graded walking test indoors on a treadmill using the modified Bruce protocol, and a submaximal 1 km outdoor uphill cardio-trekking test (1 km CTT). Heart rate (HR) and oxygen consumption (V̇O2) were continuously measured throughout both tests. BRPE was simultaneously assessed at the end of each increment on the treadmill, while the maximal BRPE value was noted at the end of the 1 km CTT. RESULTS: On the treadmill, BRPE correlated very high with relative HR (%HRmax) (ρ = 0.88, p < 0.001) and V̇O2 (%V̇O2max) (ρ = 0.89, p < 0.001). During the 1 km CTT, a small correlation between BRPE and %HRmax (ρ = 0.24, p < 0.05), respectively %V̇O2max was found (ρ = 0.24, p < 0.05). CONCLUSIONS: Criterion validity of BRPE during uphill walking depends on the environment and is higher during a treadmill test compared to a natural environment. Adding sensor-based, objective exercise-intensity parameters such as HR holds promise to improve intensity prescription and health safety during uphill walking in a natural environment.

Development and validation of a 1-km cardio-trekking test to estimate cardiorespiratory fitness in healthy adults.

Maximum oxygen uptake (V̇O2max), the gold standard measure of cardiorespiratory fitness (CRF), supports cardiovascular risk assessment and is mainly assessed during maximal spiroergometry. However, for field use, submaximal exercise tests might be appropriate and feasible. There have been no studies attempting a submaximal test protocol involving uphill hiking. This study aimed to develop and validate a 1-km cardio-trekking test (CTT) controlled by heart rate monitoring and Borg's 6-20 rating of perceived exertion (RPE) scale to predict V̇O2max outdoors. Healthy participants performed a maximal incremental treadmill walking laboratory test and a submaximal 1-km CTT on mountain trails in Austria and Germany, and V̇O2max was assessed with a portable spirometry device. Borg's RPE scale was used to control the exercise intensity of the CTT. All subjects wore a chest strap to measure heart rate (HR). A total of 134 participants (median age: 56.0 years [IQR: 51.8-63.0], 43.3 % males) completed both testing protocols. The prediction model is based on age, gender, smoking status, weight, mean HR, altitude difference, duration, and the interaction between age and duration (R2 = 0.65, adj. R2 = 0.63). Leave-one-out cross-validation revealed small shrinkage in predictive accuracy (R2 = 0.59) compared to the original model. Submaximal exercise testing using uphill hiking allows for practical estimation of V̇O2max in healthy adults. This method may allow people to engage in physical activity while monitoring their CRF to avert unnecessary cardiovascular events.

Comprehensive training load monitoring with biomarkers, performance testing, local positioning data, and questionnaires - first results from elite youth soccer.

Load management, i.e., prescribing, monitoring, and adjusting training load, is primarily aimed at preventing injury and maximizing performance. The search for objective monitoring tools to assess the external and internal load of athletes is of great interest for sports science research. In this 4-week pilot study, we assessed the feasibility and acceptance of an extensive monitoring approach using biomarkers, neuromuscular performance, and questionnaires in an elite youth soccer setting. Eight male players (mean ± SD: age: 17.0 ± 0.6 years, weight: 69.6 ± 8.2 kg, height: 177 ± 7 cm, VO2max: 62.2 ± 3.8 ml/min/kg) were monitored with a local positioning system (e.g., distance covered, sprints), biomarkers (cell-free DNA, creatine kinase), questionnaires, neuromuscular performance testing (counter-movement jump) and further strength testing (Nordic hamstring exercise, hip abduction and adduction). Feasibility was high with no substantial impact on the training routine and no adverse events such as injuries during monitoring. Adherence to the performance tests was high, but adherence to the daily questionnaires was low, and decreased across the study period. Occasional significant correlations were observed between questionnaire scores and training load data, as well as between questionnaire scores and neuromuscular performance. However, due to the small sample size, these findings should be treated with caution. These preliminary results highlight the feasibility of the approach in elite soccer, but also indicate that modifications are needed in further large-scale studies, particularly in relation to the length of the questionnaire.

Influence of 2 Digital Exercise Modules of a Multimodular System on Balance and Leg Strength Under Consideration of Use Adherence: Prospective Cohort Study.

BACKGROUND: To empower healthy aging, digital solutions embed multiple modules for physical activity, cognitive health promotion, and social engagement. Integrating new empowering technologies such as digital exercise monitoring requires assessment measures and analysis procedures, considering variable compliance of users with different modules. OBJECTIVE: This study aims to assess the influence of a tablet-based and a feedback system-based exercise module on balance and leg strength by considering use adherence instead of the use of the entire multimodular system. METHODS: In the prospective cohort study within the fit4AAL project, 83 users (n=67, 81% women; n=16, 19% men; mean age 66.2, SD 2.3 years) used the 2 digital exercise modules of a multimodular physical activity promotion system for >18 weeks. A data-driven clustering method based on the average use frequency of the exercise modules determined the number of user types that met the World Health Organization-recommended training frequency of at least twice per week. On the basis of this use adherence, statistical analysis was performed with features of functional performance tests (unipedal stance, 30-second chair rise, Y-balance, and hurdle step tests). The tests were conducted 6 months before the intervention, immediately before the intervention, and after the intervention, comparing the baseline phase with the 3 feedback use groups of the study (using only the tablet, the tablet and the feedback system, or only the feedback system). RESULTS: Of the 83 users, 43 (52%) met the World Health Organization-recommended frequency of muscle-strengthening activities. Overall, the feedback use groups achieved, on average, more chair rises in 30 seconds than the baseline group (P=.01; moderate effect size of 0.07). Of the 43 users, 26 (60%) additionally used the feedback system-based exercise module. They improved in balance compared with the users using either the tablet or the feedback system (P=.02). In addition, they improved their leg strength within the group (P=.04) and compared with the baseline (P=.01). CONCLUSIONS: The additional use of a feedback system showed a tendency to positively maintain and influence the already exceptionally high functional performance of older adults. Considering use adherence in future multimodular system studies is crucial to assess the influence of single and combined use of exercise modules on functional performance.

Data-Driven User-Type Clustering of a Physical Activity Promotion App: Usage Data Analysis Study.

BACKGROUND: Physical inactivity remains a leading risk factor for mortality worldwide. Owing to increasing sedentary behavior (activities in a reclining, seated, or lying position with low-energy expenditures), vehicle-based transport, and insufficient physical workload, the prevalence of physical activity decreases significantly with age. To promote sufficient levels of participation in physical activities, the research prototype Fit-mit-ILSE was developed with the goal of making adults aged ≥55 years physically fit and fit for the use of assistive technologies. The system combines active and assisted living technologies and smart services in the ILSE app. OBJECTIVE: The clustering of health and fitness app user types, especially in the context of active and assisted living projects, has been mainly defined by experts through 1D cluster thresholds based on app usage frequency. We aimed to investigate and present data-driven methods for clustering app user types and to identify usage patterns based on the ILSE app function Fit at home. METHODS: During the 2 phases of the field trials, ILSE app log data were collected from 165 participants. Using this data set, 2 data-driven approaches were applied for clustering to group app users who were similar to each other. First, the common approach of user-type clustering based on expert-defined thresholds was replaced by a data-driven derivation of the cluster thresholds using the Jenks natural breaks algorithm. Second, a multidimensional clustering approach using the Partitioning Around Medoids algorithm was explored to consider the detailed app usage pattern data. RESULTS: Applying the Jenks clustering algorithm to the mean usage per day and clustering the users into 4 groups showed that most of the users (63/165, 38.2%) used the Fit at home function between once a week and every second day. More men were in the low usage group than women. In addition, the younger users were more often identified as moderate or high users than the older users, who were mainly classified as low users; moreover, the regional differences between Vienna and Salzburg were identified. In addition, the multidimensional approach identified 4 different user groups that differed mainly in terms of time of use, gender, and region. Overall, the younger women living in Salzburg were the users with highest average app usage. CONCLUSIONS: The application of different clustering approaches showed that data-driven calculations of user groups can complement expert-based definitions, provide objective thresholds for the analysis of app usage data, and identify groups that can be targeted individually based on their specific group characteristics.

Valorization of Natural Cardio Trekking Trails Through Open Innovation for the Promotion of Sustainable Cross-generational Health-Oriented Tourism in the Connect2Move Project: Protocol for a Cross-sectional Study.

BACKGROUND: Hiking is one of the most popular forms of exercise in the alpine region. However, besides its health benefits, hiking is the alpine activity with the highest incidence of cardiac events. Most incidents occur due to overexertion or underestimation of the physiological strain of hiking. OBJECTIVE: This project will establish a standardized cardio trekking test trail to evaluate the exercise capacity of tourists within hiking areas and deliver a tool for the prevention of hiking-associated cardiac incidents. Further, individual exercise intensity for a hiking tour will be predicted and visualized in digital maps. METHODS: This cooperation study between Austria and Germany will first validate a 1-km outdoor cardio trekking test trail at 2 different study sites. Then, exercise intensity measures on 8-km hiking trails will be evaluated during hiking to estimate overall hiking intensity. A total of 144 healthy adults (aged >45 years) will perform a treadmill test in the laboratory and a 1-km hiking test outdoors. They will wear a portable spirometry device that measures gas exchange, as well as heart rate, walking speed, ventilation, GPS location, and altitude throughout the tests. Estimation models for exercise capacity based on measured parameters will be calculated. RESULTS: The project "Connect2Move" was funded in December 2019 by the European Regional Development Fund (INTERREG V-A Programme Austria-Bavaria - 2014-2020; Project Number AB296). "Connect2Move" started in January 2020 and runs until the end of June 2022. By the end of April 2022, 162 participants were tested in the laboratory, and of these, 144 were tested outdoors. The data analysis will be completed by the end of June 2022, and results are expected to be published by the end of 2022. CONCLUSIONS: Individual prediction of exercise capacity in healthy individuals with interest in hiking aims at the prevention of hiking-associated cardiovascular events caused by overexertion. Integration of a mathematical equation into existing hiking apps will allow individual hiking route recommendations derived from individual performance on a standardized cardio trekking test trail. TRIAL REGISTRATION: ClinicalTrails.gov NCT05226806; https://clinicaltrials.gov/ct2/show/NCT05226806. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/39038.

Effects of an App-Based Physical Exercise Program on Selected Parameters of Physical Fitness of Females in Retirement: A Randomized Controlled Trial.

Modern technologies enable new options in the delivery of physical exercise programs. Specially designed app-based programs can be used to help older people in particular to integrate physical exercise into their daily lives. This study examines the influence of an app-based physical exercise program on selected parameters of physical fitness, such as muscular strength, balance, and flexibility. The women (n = 110) were on average 65.3 (± 1.5) years old and, compared to age-specific norm values, healthy. The 14-week intervention consisted of an app-based, unsupervised physical exercise program, in which the exercise frequency and duration of sessions were self-selected. The physical exercise program consisted of simple, functional exercises such as arm circles, squats, lateral raises. The participants were provided with an elastic resistance band and an exercise ball allowing them to increase exercise intensity if needed. Participants were randomly assigned to intervention group (IG) and control group (CG). 71% of the IG used the physical exercise program at least 1.2 times per week, whereas 25% of the IG showed usage rates above four times per week. Significant effects were found in the domains of muscular strength and flexibility. While IG could maintain their performance in isometric muscular strength tests and increased their flexibility, CG faced a decrease in those parameters. Thus, this app-based physical exercise program had positively influenced muscular strength and flexibility in women over 60 years of age.

Foot Strike Angle Prediction and Pattern Classification Using LoadsolTM Wearable Sensors: A Comparison of Machine Learning Techniques.

The foot strike pattern performed during running is an important variable for runners, performance practitioners, and industry specialists. Versatile, wearable sensors may provide foot strike information while encouraging the collection of diverse information during ecological running. The purpose of the current study was to predict foot strike angle and classify foot strike pattern from LoadsolTM wearable pressure insoles using three machine learning techniques (multiple linear regression-MR, conditional inference tree-TREE, and random forest-FRST). Model performance was assessed using three-dimensional kinematics as a ground-truth measure. The prediction-model accuracy was similar for the regression, inference tree, and random forest models (RMSE: MR = 5.16°, TREE = 4.85°, FRST = 3.65°; MAPE: MR = 0.32°, TREE = 0.45°, FRST = 0.33°), though the regression and random forest models boasted lower maximum precision (13.75° and 14.3°, respectively) than the inference tree (19.02°). The classification performance was above 90% for all models (MR = 90.4%, TREE = 93.9%, and FRST = 94.1%). There was an increased tendency to misclassify mid foot strike patterns in all models, which may be improved with the inclusion of more mid foot steps during model training. Ultimately, wearable pressure insoles in combination with simple machine learning techniques can be used to predict and classify a runner's foot strike with sufficient accuracy.