Browsing by Author "Ferraz, Ricardo"
Now showing 1 - 6 of 6
Results Per Page
Sort Options
- Classification of recovery states in U15, U17, and U19 sub-elite football players: a machine learning approachPublication . Teixeira, José E.; Encarnação, Samuel; Branquinho., L; Ferraz, Ricardo; Portella, Daniel L.; Monteiro, Diogo; Morgans, Ryland; M. Barbosa, Tiago; Monteiro, António Miguel; Forte, PedroIntroduction: A promising approach to optimizing recovery in youth football has been the use of machine learning (ML) models to predict recovery states and prevent mental fatigue. This research investigates the application of ML models in classifying male young football players aged under (U)15, U17, and U19 according to their recovery state. Weekly training load data were systematically monitored across three age groups throughout the initial month of the 2019–2020 competitive season, covering 18 training sessions and 120 observation instances. Outfield players were tracked using portable 18-Hz global positioning system (GPS) devices, while heart rate (HR) was measured using 1 Hz telemetry HR bands. The rating of perceived exertion (RPE 6–20) and total quality recovery (TQR 6–20) scores were employed to evaluate perceived exertion, internal training load, and recovery state, respectively. Data preprocessing involved handling missing values, normalization, and feature selection using correlation coefficients and a random forest (RF) classifier. Five ML algorithms [K-nearest neighbors (KNN), extreme gradient boosting (XGBoost), support vector machine (SVM), RF, and decision tree (DT)] were assessed for classification performance. The K-fold method was employed to cross-validate the ML outputs. Results: A high accuracy for this ML classification model (73–100%) was verified. The feature selection highlighted critical variables, and we implemented the ML algorithms considering a panel of 9 variables (U15, U19, body mass, accelerations, decelerations, training weeks, sprint distance, and RPE). These features were included according to their percentage of importance (3–18%). The results were cross-validated with good accuracy across 5-fold (79%). Conclusion: The five ML models, in combination with weekly data, demonstrated the efficacy of wearable device-collected features as an efficient combination in predicting football players’ recovery states.
- Comparison of force variables and dynamic strength index between age groups in elite young Brazilian football goalkeepersPublication . Portella, Daniel Leite; Jatene, Pedro; Da Silva, Alex O.; dos Santos, Gustavo S.; Monteiro, Diogo; Teixeira, José E.; Branquinho, Luís; Ferraz, Ricardo; Forte, PedroIntroduction: The application of muscle force is a determinant of football success as it is inherent to the motor control and sport. The aims of this study are: (1) to describe force variables Isometric Maximal Force (IMF), Concentric Peak Force (CPPF), and Dynamic Strength Index (DSI) in football goalkeepers from different age groups; (2) to compare these variables’ behavior between those groups. Methods: The sample was formed by 19 youth players (15.97 ± 1.55 years old) from a first-division Brazilian football team. The CPPF and IMF variables were obtained through the Countermovement jump and isometric squat tests, respectively. For data collection, a force plate (Cefise, Brazil) was used with an acquisition frequency of 600 Hz and mono-axial. The DSI was calculated using the ratio between CPPF and IMF. For data analysis, the sample was separated into clusters by age. After the grouping, a descriptive analysis of the data and a comparison between the groups with p < 0.05. Results: The sample was grouped into three groups (GA, GB, and GC) and one of the individuals did not enter the group, totaling 18 individuals in the analyzed sample. The comparison between the ages of the groups showed a significant difference and small and moderate effect size (ES), validating the cluster strategy. The CPPF and IMF variables showed increased values according to chronological age. CPPF showed a significant difference between GA-GB, (ES = very large) GA-GC (ES = very large), and GB-GC (ES = moderate). The IMF variable had significant differences between GA-GB (ES = moderate) and GA-GC (ES = very large). However, DSI showed significant differences only between GA GB (ES = small) and GB-GC (ES = very large). Conclusions: The CPPF and IMF variables had constant increases and distinct values with an increase according to age, and this did not occur for DSI. The difference between CPPF and IMF compared to DSI bring to light the variability in dynamics and proportionality between muscular force in the concentric phase and maximal force in the isometric regime during the developmental process over chronological age in soccer goalkeepers.
- A Deep Learning Neural Network to Classify Obesity Risk in Portuguese Adolescents Based on Physical Fitness Levels and Body Mass Index Percentiles: Insights for National Health PoliciesPublication . Forte, Pedro; Encarnação, Samuel; Monteiro, António; Teixeira, José; Hattabi, Soukaina; Sortwell, Andrew; Branquinho, Luís; Amaro, Bruna; Sampaio, Tatiana; Flores, Pedro; Silva-santos, Sandra; Ribeiro, Joana; Batista, Amanda; Ferraz, Ricardo; Rodrigues, FilipeThe increasing prevalence of overweight and obesity among adults is a risk factor for many chronic diseases and death. In addition, obesity among children and adolescents has reached unprecedented levels and studies show that obese children and adolescents are more likely to become obese adults. Therefore, both the prevention and treatment of obesity in adolescents are critical. This study aimed to develop an artificial intelligence (AI) neural network (NNET) model that identifies the risk of obesity in Portuguese adolescents based on their body mass index (BMI) percentiles and levels of physical fitness. Using datasets from the FITescola® project, 654 adolescents aged between 10–19 years old, male: 334 (51%), female: n = 320 (49%), age 13.8 ± 2 years old, were selected to participate in a cross-sectional observational study. Physical fitness variables, age, and sex were used to identify the risk of obesity. The NNET had good accuracy (75%) and performance validation through the Receiver Operating Characteristic using the Area Under the Curve (ROC AUC = 64%) in identifying the risk of obesity in Portuguese adolescents based on the BMI percentiles. Correlations of moderate effect size were perceived for aerobic fitness (AF), upper limbs strength (ULS), and sprint time (ST), showing that some physical fitness variables contributed to the obesity risk of the adolescents. Our NNET presented a good accuracy (75%) and was validated with the K-Folds Cross-Validation (K-Folds CV) with good accuracy (71%) and ROC AUC (66%). According to the NNET, there was an increased risk of obesity linked to low physical fitness in Portuguese teenagers.
- Obesity Status and Physical Fitness Levels in Male and Female Portuguese Adolescents: A Two-Way Multivariate AnalysisPublication . Encarnação, Samuel; Rodrigues, Filipe; Monteiro, António; Gouili, Hatem; Hattabi, Soukaina; Sortwell, Andrew; Branquinho, Luís; Teixeira, José; Ferraz, Ricardo; Flores, Pedro; Silva-Santos, Sandra; Ribeiro, Joana; Batista, Amanda; Forte, PedroObesity and decreasing fitness levels among the youth are growing concerns in Portugal, similar to other developed countries, with implications for health and psychomotor development. Understanding the influence of health determinants such as sex and age are crucial for developing effective public health strategies. This study aimed to analyze the association between sex and chronological age with obesity status and physical fitness in Portuguese adolescents. A total of 170 adolescents (85 males and 85 females) were evaluated for body mass index, abdominal adiposity, aerobic fitness, abdominal resistance, upper limb resistance, lower limb power, and maximal running speed in a 40 m sprint using the FITescola® physical fitness battery, a Portuguese government initiative. The general model, analyzed using Pillai’s trace, showed a significant effect of age and sex on body mass index, abdominal circumference, aerobic fitness, abdominal resistance, upper limb resistance, lower limb power, and maximal running speed (V = 0.99, F (7) = 10,916.4, p < 0.001, partial η2, sex = 0.22; age = 0.43, sex and age interaction = 0.10). Boys had higher physical fitness levels than girls in most tests, but both sex groups had a significantly higher proportion of non-fit adolescents, with boys showing the highest number of participants classified as non-fit.
- Physical Activity and Sedentary Behavior in Adults from Penafiel, Portugal: A Cross-Sectional StudyPublication . Batista, Amanda; Forte, Pedro; Ribeiro, Joana; Silva-Santos, Sandra; Neto, Elmiro; Rodrigues, Filipe; Teixeira, José; Ferraz, Ricardo; Branquinho, LuísThe aim of this study was to compare the physical activity (PA) and sedentary behavior (SB) levels of young and middle-aged adults living in and around the municipality of Penafiel and to determine whether they meet PA recommendations. The researchers used the “International Physical Activity Questionnaire” (IPAQ) to measure moderate to vigorous PA and time spent on sedentary behavior (high vs. low). A prospective observational cross-sectional sample of 1105 adults aged 18–63 years, living in the municipality of Penafiel and its surroundings (45% women, 55% men), was used. The results indicated that more than half of the population was inactive (53.8%) and sedentary (54.0%). Men were more likely to be sedentary (59.2%) and inactive (55.6%) than women (inactive: 51.7%, high SB: 47.7%). Regarding daily PA and SB levels, women had higher levels of walks (3.8 ± 2.3; p = 0.034) and vigorous PA (2.2 ± 1.8 min; p = 0.005) per days/week, as well as vigorous PA per minutes/week (75.4 ± 82.1 min; p = 0.034). The time spent on vigorous PA per day was also higher in women (26.2 ± 22.8 min; p = 0.030). However, men had higher values in walking minutes per day (26.3 ± 17.1 min; p = 0.030), SB for weekdays (429.2 ± 141.2 min; p = 0.001), SB for weekends (324.7 ± 163.7 min; p = 0.033) and time spent on SB per minutes/week (2795.6 ± 882.0 min; p = 0.001). The results also showed that the older the adults, the lower the frequency and total time of vigorous PA per week. Young adults (18–28 years) had higher levels of vigorous PA (p = 0.005) than the other age groups (29–39; 40–50 and 51–63 years). Finally, the study found no significant correlation between individual level factors, such as number of children, marital status and monthly income, and PA or SB. Conversely, a significant and negative correlation between SB and levels of PA was found, indicating that the higher the level of PA practice, the lower the SB level. The authors suggest that promoting new PA habits and healthy lifestyles is an important future challenge for sustainability and improving the quality of life in public health.
- The Association between Training Frequency, Symptoms of Overtraining and Injuries in Young Men Soccer PlayersPublication . Rodrigues, Filipe; Monteiro, Diogo; Ferraz, Ricardo; Branquinho, Luís; Forte, PedroOvertraining is a prevalent issue among young men soccer players, particularly those who are driven to enhance their skills. While an intense training volume and effort might contribute to athletic growth, it can also have negative implications, including injury. The current study aimed at examining the association between training frequency, symptoms of overtraining and injuries in young men soccer players. A path analysis approach was used to examine the causal relationships between variables. The sample consisted of 189 young men soccer players aged 13–17 years old (age = 14.81, SD = 1.37). Participants reported that they were training, on average, 5.77 days (SD = 1.53) per week. Athletes were competing at a regional (n = 100) or national (n = 89) level. Concerning injuries, participants indicated, on average, 2.03 (SD = 1.16) injuries since they started practicing soccer. The results displayed a significant association, as theoretically expected, namely: (i) training frequency was significantly associated with overtraining symptoms (β = 0.15 [IC95% = 0.01, 0.29]); (ii) overtraining symptoms were significantly associated with the number of injuries (β = 0.19 [IC95% = 0.02, 0.35]). An indirect effect between training frequency and injuries was also observed (β = 0.15 [IC95% = 0.01, 0.29]). Thus, there is preliminary evidence that overtraining symptoms could play a mediating role. In conclusion, investigating the links between overtraining symptoms and injury in young men soccer players is critical, as it can assist in identifying overtraining warning signs, promote young players’ health and safety, customize training regimens to individual needs, and contribute to a better understanding of sports-related injuries.