Machine Learning on Prediction of Relative Physical Activity Intensity Using Medical Radar Sensor and 3D Accelerometer.

BACKGROUND: One of the most critical topics in sports safety today is the reduction in injury risks through controlled fatigue using non-invasive athlete monitoring. Due to the risk of injuries, it is prohibited to use accelerometer-based smart trackers, activity measurement bracelets, and smart watches for recording health parameters during performance sports activities. This study analyzes the synergy feasibility of medical radar sensors and tri-axial acceleration sensor data to predict physical activity key performance indexes in performance sports by using machine learning (ML). The novelty of this method is that it uses a 24 GHz Doppler radar sensor to detect vital signs such as the heartbeat and breathing without touching the person and to predict the intensity of physical activity, combined with the acceleration data from 3D accelerometers. METHODS: This study is based on the data collected from professional athletes and freely available datasets created for research purposes. A combination of sensor data management was used: a medical radar sensor with no-contact remote sensing to measure the heart rate (HR) and 3D acceleration to measure the velocity of the activity. Various advanced ML methods and models were employed on the top of sensors to analyze the vital parameters and predict the health activity key performance indexes. three-axial acceleration, heart rate data, age, as well as activity level variances. RESULTS: The ML models recognized the physical activity intensity and estimated the energy expenditure on a realistic level. Leave-one-out (LOO) cross-validation (CV), as well as out-of-sample testing (OST) methods, have been used to evaluate the level of accuracy in activity intensity prediction. The energy expenditure prediction with three-axial accelerometer sensors by using linear regression provided 97-99% accuracy on selected sports (cycling, running, and soccer). The ML-based RPE results using medical radar sensors on a time-series heart rate (HR) dataset varied between 90 and 96% accuracy. The expected level of accuracy was examined with different models. The average accuracy for all the models (RPE and METs) and setups was higher than 90%. CONCLUSIONS: The ML models that classify the rating of the perceived exertion and the metabolic equivalent of tasks perform consistently.

Analysis of different training load monitoring methods in youth women handball players

This study examined the association between different methods for training load (TL) monitoring during youth handball training. Distance covered, heart rate and session rating of perceived exertion [SRPE] were recorded during 12 training sessions in 14 youth women handball athletes (16.9±1.1 years). Internal load models based on SRPE and Edwards’ Trimp were calculated. An oscillatory feature was observed for the three methods of TL assessment (SRPE: 383±159 A.U., Edwards’ Trimp: 252±71 A.U., total distance: 3997±1291 m). A large correlation was found between Edwards’ Trimp and distance covered (r=0.59). A moderate correlation was observed for Edwards’ Trimp vs. SRPE (r=0.36), and between SRPE vs. distance covered (r=0.49). Shared variances of 13–35% were observed between TL methods comparisons. The results suggest that different constructs seem to be measured by each load model. Additionally, SRPE is a simple and low-cost method that might be used for TL monitoring in handball. (AU)

Internal Training Load Perceived by Athletes and Planned by Coaches: A Systematic Review and Meta-Analysis.

BACKGROUND: Rating of perceived exertion (RPE) and session RPE (sRPE) has been widely used to verify the internal load in athletes. Understanding the agreement between the training load prescribed by coaches and that perceived by athletes is a topic of great interest in sport science. OBJECTIVE: This systematic review and meta-analysis aimed to investigate differences between the training/competition load perceived by athletes and prescribed/intended/observed by coaches. METHODS: A literature search (September 2020 and updated in November 2021) was conducted using PubMed, Web of Science, Embase, and SPORTDiscus databases. The protocol was registered in the Open Science Framework (osf.io/wna4x). Studies should include athletes and coaches of any sex, age, or level of experience. The studies should present outcomes related to the RPE or sRPE for any scale considering overall training/competition sessions (physical, strength, tactical, technical, games) and/or classified into three effort categories: easy, moderate, and hard. RESULTS: Twenty-seven studies were included in the meta-analysis. No difference was found between coaches and athletes for overall RPE (SMD = 0.19, P = 0.10) and overall sRPE (SMD = 0.05, P = 0.75). There was a difference for easy RPE (SMD = - 0.44, small effect size, P = 0.04) and easy sRPE (SMD = - 0.54, moderate effect size, P = 0.04). No differences were found for moderate RPE (SMD = 0.05, P = 0.74) and hard RPE (SMD = 0.41, P = 0.18). No difference was found for moderate (SMD = -0.15, P = 0.56) and hard (SMD = 0.20, P = 0.43) sRPE. CONCLUSION: There is an agreement between coaches and athletes about overall RPE and sRPE, and RPE and sRPE into two effort categories (moderate and hard). However, there were disagreements in RPE and sRPE for easy effort category. Thus, despite a small disagreement, the use of these tools seems to be adequate for training monitoring.

Low-Volume Sprint Interval Swimming Is Sufficient to Increase Blood Metabolic Biomarkers in Master Swimmers.

Purpose: Sprint interval exercise is a time-efficient way of inducing beneficial adaptations. However, little is known about its minimal effective volume, especially in swimming. The aim of the present study was to evaluate and compare the effects of two sprint interval swimming sets of different low volumes on blood biomarkers. Method: Twenty-one master swimmers [11 females aged 38.5 (8.5) years, 10 males aged 42.7 (5.7) years] completed two freestyle swimming sets of 4 × 50 m and 4 × 25 m at maximal intensity and a work-to-rest ratio of 1:1, on different days, in random and counterbalanced order. Blood samples were taken before, immediately after and one hour after exercise for determination of a number of biochemical parameters. Results: Swimming speed was higher in the 4 × 25-m set. Lactate, glucose, insulin, glucagon, cortisol, and reduced glutathione increased immediately post-exercise, while uric acid increased 1 h post-exercise (p < .05). All aforementioned biomarkers, excluding glucagon, increased more with the 4 × 50-m set, compared to the 4 × 25-m set (p < .05). Session rating of perceived exertion was higher after the 4 × 50-m set (p = .011). Conclusion: Both sprint interval swimming sets elicited increases in blood biomarkers in master swimmers. The set of higher volume elicited greater increases in most of the biochemical markers studied but also in subjective load. Thus, although the set of higher volume was more efficient in perturbing blood biomarkers, even the very low-volume set induced metabolic stress that may trigger adaptive mechanisms.

Association of the acute:chronic workload ratio and wellness scores in premier league male hockey players.

Background: The relationship between acute:chronic workload ratios (ACWR) and the incidence of injury, as well as the relationship between subjective wellness scores and training load, is unclear in hockey players. Objectives: This study investigated these relationships to determine if the ACWR is a useful workload management tool for field hockey players. In addition, the study investigated the association between specific subjective wellness scores (fatigue, sleep quality, general muscle soreness, mood and stress level) and the acute:chronic workload ratio and training load. The study also assessed the association between individual players' training loads with the self-reported wellness scores. Methods: Fourteen male field hockey players from the men's first team at the University of the Witwatersrand participated in the study over ten weeks during a premier league competition phase. A Google form was completed within thirty minutes of every training session/match. This form consisted of questions that allowed for the calculation of the ACWR. Results: No incidence of injury was reported during the ten-week period. ACWR scores ranged between 0.67 and 1.87. The ACWR was associated with general muscle soreness (p = 0.010) and training load was associated with fatigue (p = 0.002), sleep quality (p = 0.05), general muscle soreness (p = 0.004), and mood (p=0.025). Conclusion: There may be some merit in the use of subjective wellness measures as workload management tools in field hockey. Further research is required to determine if there is an optimal ACWR for this sport.

Training Load and Recovery During a Pre-Olympic Season in Professional Rhythmic Gymnasts.

CONTEXT: Rhythmic gymnastics requires a high level of complexity and perfection of technical gestures, associated with well-developed physical and artistic capacities. The training-load and recovery profiles of rhythmic gymnasts across a season are unknown. OBJECTIVE: To analyze the training load and recovery of professional rhythmic gymnasts during 1 season. DESIGN: Cohort study. SETTING: Brazilian National Training Center of Rhythmic Gymnastics and competition facilities. PATIENTS OR OTHER PARTICIPANTS: Eight gymnasts from the Brazilian national senior rhythmic gymnastics group. MAIN OUTCOME MEASURE(S): Session rating of perceived exertion (session-RPE) and total quality recovery (TQR) scores were collected daily for 43 weeks. We obtained the session-RPE after each session and TQR score before the first session of the day. Performances during 5 competitions were also recorded. The season was divided into 8 periods. Total weekly internal training load (wITL), training intensity, frequency, duration, recovery, and acute : chronic workload ratio were calculated for analysis. RESULTS: The season mean wITL was 10 381 ± 4894 arbitrary units, mean session-RPE score was 5.0 ± 1.6, and mean TQR score was 12.8 ± 1.3. The gymnasts trained an average of 8.7 ± 2.9 sessions per week, with a mean duration of 219 ± 36 minutes. Each competitive period showed increased wITL compared with the previous period. Training-load variables (wITL and session-RPE) and recovery were inversely correlated. Gymnasts were poorly recovered (TQR < 13) during 50.9% of the season (n = 167 times), especially during competitive weeks. Spikes in load (acute : chronic workload ratio ≥ 1.5) occurred across 18.1% of the season (n = 55 times). CONCLUSIONS: The training-load variables and recovery changed throughout a professional rhythmic gymnastics group season, mainly during competitive periods. The correct distribution of training load is critical to ensure that gymnasts are entering competitions in a recovered state.

Workloads of forward and backline adolescent rugby players: a pilot study.

Background: There is minimal research on workloads of adolescent rugby players. Therefore, the main aim of this study was to determine the workloads placed on a cohort of South African adolescent rugby players (n = 17), during an in-season period. Methods: Session RPE ratings were collected daily, 30 minutes after the training session concluded, during an 11-week in-season period. The training load was calculated as the session ratings of perceived exertion multiplied by the session's duration (min). Results: The main finding of the study was that the adolescents in this investigation had similar workloads to elite players but higher workloads than other studies on adolescent rugby players. The forwards (3311±939 arbitrary units; AU) had a higher workload than backline players (2851±1080 AU). There was no difference between forwards and backline players with regards to the acute:chronic workload ratio. Conclusion: Workloads are high in these adolescent players, particularly in the forwards, and are similar to the workloads of elite level rugby players.

Combined Small-Sided Game and High-Intensity Interval Training in Soccer Players: The Effect of Exercise Order.

The aim of the present study was to compare combined small-sided game (SSG) and high-intensity interval training (HIT) with different order. Twenty-one semi-professional soccer players were divided into two groups: SSG+HIT (n = 10) and HIT+SSG (n = 11), and underwent similar four-week training programs. Players completed the 30-15 Intermittent Fitness Test (30-15IFT) before and after the experiment; maximum speed (VIFT) was recorded. During the experiment, seven sessions of SSG (3 vs 3) and HIT (15"-15" with 95-100% VIFT) were implemented. Weekly accumulated training loads for both groups during the experiment were similar. Moderate improvements in VIFT were observed in both SSG+HIT (+6.2%, 90% confidence limits, [CL] 4.6; 7.7 and Effect Size, [ES] +0.96) and HIT+SSG (+6.9%, 90% CL 4.6; 9.3 and ES +0.97) groups. Between-group difference in changes of VIFT was trivial (+0.7%, 90% CL -1.8; 3.3 and ES +0.11). Combining SSG and HIT in different order elicited the same enhancement in high-intensity intermittent performance in soccer players.

Convergent validity of a novel method for quantifying rowing training loads.

Elite rowers complete rowing-specific and non-specific training, incorporating continuous and interval-like efforts spanning the intensity spectrum. However, established training load measures are unsuitable for use in some modes and intensities. Consequently, a new measure known as the T2minute method was created. The method quantifies load as the time spent in a range of training zones (time-in-zone), multiplied by intensity- and mode-specific weighting factors that scale the relative stress of different intensities and modes to the demands of on-water rowing. The purpose of this study was to examine the convergent validity of the T2minute method with Banister's training impulse (TRIMP), Lucia's TRIMP and Session-RPE when quantifying elite rowing training. Fourteen elite rowers (12 males, 2 females) were monitored during four weeks of routine training. Unadjusted T2minute loads (using coaches' estimates of time-in-zone) demonstrated moderate-to-strong correlations with Banister's TRIMP, Lucia's TRIMP and Session-RPE (rho: 0.58, 0.55 and 0.42, respectively). Adjusting T2minute loads by using actual time-in-zone data resulted in stronger correlations between the T2minute method and Banister's TRIMP and Lucia's TRIMP (rho: 0.85 and 0.81, respectively). The T2minute method is an appropriate in-field measure of elite rowing training loads, particularly when actual time-in-zone values are used to quantify load.