Dataset Information

Essential elements of physical fitness analysis in male adolescent athletes using machine learning.

ABSTRACT: Physical fitness (PF) includes various factors that significantly impacts athletic performance. Analyzing PF is critical in developing customized training methods for athletes based on the sports in which they compete. Previous approaches to analyzing PF have relied on statistical or machine learning algorithms that focus on predicting athlete injury or performance. In this study, six machine learning algorithms were used to analyze the PF of 1,489 male adolescent athletes across five sports, including track & field, football, baseball, swimming, and badminton. Furthermore, the machine learning models were utilized to analyze the essential elements of PF using feature importance of XGBoost, and SHAP values. As a result, XGBoost represents the highest performance, with an average accuracy of 90.14, an area under the curve of 0.86, and F1-score of 0.87, demonstrating the similarity between the sports. Feature importance of XGBoost, and SHAP value provided a quantitative assessment of the relative importance of PF in sports by comparing two sports within each of the five sports. This analysis is expected to be useful in analyzing the essential PF elements of athletes in various sports and recommending personalized exercise methods accordingly.

SUBMITTER: Lee YH

PROVIDER: S-EPMC10986970 | biostudies-literature | 2024

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Essential elements of physical fitness analysis in male adolescent athletes using machine learning.

Lee Yun-Hwan YH Chang Jisuk J Lee Ji-Eun JE Jung Yeon-Sung YS Lee Dongheon D Lee Ho-Seong HS

PloS one 20240402 4

Physical fitness (PF) includes various factors that significantly impacts athletic performance. Analyzing PF is critical in developing customized training methods for athletes based on the sports in which they compete. Previous approaches to analyzing PF have relied on statistical or machine learning algorithms that focus on predicting athlete injury or performance. In this study, six machine learning algorithms were used to analyze the PF of 1,489 male adolescent athletes across five sports, in ...[more]

PMID: 38564629

Similar Datasets

Project description:Sport concussions can be difficult to diagnose and if missed, they can expose athletes to greater injury risk and long-lasting neurological disabilities. Discovery of objective biomarkers to aid concussion diagnosis is critical to protecting athlete brain health. To this end, we performed targeted proteomics on plasma obtained from adolescent athletes suffering a sports concussion. A total of 11 concussed male athletes were enrolled at our academic Sport Medicine Concussion Clinic, as well as 24 sex-, age- and activity-matched healthy control subjects. Clinical evaluation was performed and blood was drawn within 72 h of injury. Proximity extension assays were performed for 1,472 plasma proteins; a total of six proteins were considered significantly different between cohorts (P < 0.01; five proteins decreased and one protein increased). Receiver operating characteristic curves on the six individual protein biomarkers identified had areas-under-the-curves (AUCs) for concussion diagnosis ≥0.78; antioxidant 1 copper chaperone (ATOX1; AUC 0.81, P = 0.003), secreted protein acidic and rich in cysteine (SPARC; AUC 0.81, P = 0.004), cluster of differentiation 34 (CD34; AUC 0.79, P = 0.006), polyglutamine binding protein 1 (PQBP1; AUC 0.78, P = 0.008), insulin-like growth factor-binding protein-like 1 (IGFBPL1; AUC 0.78, P = 0.008) and cytosolic 5'-nucleotidase 3A (NT5C3A; AUC 0.78, P = 0.009). Combining three of the protein biomarkers (ATOX1, SPARC and NT5C3A), produced an AUC of 0.98 for concussion diagnoses (P < 0.001; 95% CI: 0.95, 1.00). Despite a paucity of studies on these three identified proteins, the available evidence points to their roles in modulating tissue inflammation and regulating integrity of the cerebral microvasculature. Taken together, our exploratory data suggest that three or less novel proteins, which are amenable to a point-of-care immunoassay, may be future candidate biomarkers for screening adolescent sport concussion. Validation with protein assays is required in larger cohorts.

Project description:BackgroundThe identification of sport talent among adolescent athletes is a topic that in recent years has been a major focus of interest for both the scientific community and sport managers. Both anthropometry and physical performance through fitness tests have demonstrated to be key elements. Biological maturation, due to its influence on anthropometric variables and physical fitness, has also been studied in relation to sport talent identification.ObjectiveTo analyse differences according to biological maturation status in anthropometric characteristics and performance in physical fitness tests, and to determine which variables predict better performance in physical fitness tests in adolescent volleyball players.MethodsA cross-sectional design was followed to collect the data. A total of 48 male sub-elite volleyball players (14.17 ± 0.73 years) completed a socio-demographic and sports ad hoc questionnaire. Anthropometric variables were measured following the guidelines of the International Society for the Advancement in Kinanthropometry (ISAK) including four basic measurements (body mass, height, sitting height and arm span); eight skinfolds (triceps, biceps, subscapular, iliac crest, supraspinale, abdominal, thigh and calf); four girths (arm relaxed, flexed and tensed arm, middle thigh and calf); five breadths (biacromial, biileocrestal, humerus, femur and bi-styloid); three lengths (acromiale-radiale, radiale-stylion and stylion-medio dactilion); and a height (ilioespinale). Physical fitness was assessed, including the sit-and-reach, back scratch, long jump, medicine ball throw, counter movement jump (CMJ), 20 meters sprint, and agility tests. Furthermore, maturity offset and age at peak height velocity (APHV) was calculated.ResultsSignificant differences were found in the body mass (Mean Difference, MD = 20.86-30.75), height (MD = 11.72-19.09), sitting height (MD = 4.27-10.27), arm span (MD = 12.91-20.78), body mass index (MD = 3.72-5.63), upper limb length (MD = 7.76), corrected muscle girths (MD = 2.06-9.31), ∑6 and 8 skinfolds (MD = 3.67-50.21) fat mass and percentage (MD = 0.30-11.58), muscle (MD = 4.13-10.64) and bone mass (MD = 1.61-3.54) (p < 0.001-0.030), showing higher values the early maturers. In the physical fitness tests, significant differences were observed in the medicine ball throw (MD = 1.26-2.80) and in CMJ power (MD = 156.71-379.85) (p < 0.001). Regression models identified fat mass percentage predicted worse physical test performance (p < 0.001), while age, maturation offset, muscle and bone variables were predictors of better physical performance (p < 0.001).ConclusionsSignificant differences based upon the stages of biological maturation were found in the anthropometric and physical condition variables in favor of the players whose maturation process was more advanced, with the variables related to fat and adipose, muscle and bone development conditioning their performance in the physical condition tests.

Dataset Information

Essential elements of physical fitness analysis in male adolescent athletes using machine learning.

Publications

Essential elements of physical fitness analysis in male adolescent athletes using machine learning.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets