Health science reports · 2026
Wearable Technologies and Artificial Intelligence in Sports Medicine: A Systematic Narrative Review of Injury Prevention and Rehabilitation.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Silva Margarida, Al-Jumaili Saif, Lopes Mário, Albuquerque Carlos, Reis António D, Coelho Paulo Jorge, Pires Ivan Miguel
Аннотация
BACKGROUND AND AIMS: Wearable technologies and artificial intelligence (AI) are shifting sports medicine from episodic, subjective assessment toward continuous, data-driven injury surveillance. This review aimed to map the main wearable sensor modalities used in injury prevention and rehabilitation, evaluate AI architectures applied to biomechanical and physiological data, summarize clinically relevant applications by anatomical region, and identify methodological, ethical, and implementation challenges.
METHODS: A systematic narrative review was conducted across PubMed, Scopus, Web of Science, and Google Scholar for studies published from 2016 to 2026, supplemented by manual reference screening. Search terms combined wearable sensors, AI and machine learning, and sports medicine outcomes. After title and abstract screening of 614 records and full-text assessment of 87 articles, 28 studies met the inclusion criteria.
RESULTS: Inertial measurement units, GPS/GNSS, surface electromyography, and photoplethysmography were the most frequently reported sensor modalities. Deep learning, ensemble methods, and multimodal fusion were commonly used to process biomechanical and physiological signals. One hybrid IMU-sEMG study involving 50 athletes reported 92.3% accuracy, 90.5% recall, and an area under the receiver operating characteristic curve of 0.93 for classifying biomechanical injury-risk states. An ankle-monitoring study using data from 87 athletes across 12 sports reported 87.4% overall detection accuracy, a 12.1% false-alarm rate, and 17.2 ms latency. However, substantial heterogeneity in study design, populations, outcomes, and validation methods prevented direct comparison or pooled quantitative interpretation.
CONCLUSION: Wearable-AI systems show promising potential for injury-risk monitoring, rehabilitation, and return-to-play support. Wider clinical adoption will require more diverse populations, longitudinal and field-based validation, improved model interpretability, standardized protocols, and robust ethical, legal, and governance frameworks.
METHODS: A systematic narrative review was conducted across PubMed, Scopus, Web of Science, and Google Scholar for studies published from 2016 to 2026, supplemented by manual reference screening. Search terms combined wearable sensors, AI and machine learning, and sports medicine outcomes. After title and abstract screening of 614 records and full-text assessment of 87 articles, 28 studies met the inclusion criteria.
RESULTS: Inertial measurement units, GPS/GNSS, surface electromyography, and photoplethysmography were the most frequently reported sensor modalities. Deep learning, ensemble methods, and multimodal fusion were commonly used to process biomechanical and physiological signals. One hybrid IMU-sEMG study involving 50 athletes reported 92.3% accuracy, 90.5% recall, and an area under the receiver operating characteristic curve of 0.93 for classifying biomechanical injury-risk states. An ankle-monitoring study using data from 87 athletes across 12 sports reported 87.4% overall detection accuracy, a 12.1% false-alarm rate, and 17.2 ms latency. However, substantial heterogeneity in study design, populations, outcomes, and validation methods prevented direct comparison or pooled quantitative interpretation.
CONCLUSION: Wearable-AI systems show promising potential for injury-risk monitoring, rehabilitation, and return-to-play support. Wider clinical adoption will require more diverse populations, longitudinal and field-based validation, improved model interpretability, standardized protocols, and robust ethical, legal, and governance frameworks.