Use of machine learning and deep learning for exercise prescription
DOI:
https://doi.org/10.56294/dm20251054Keywords:
Artificial Intelligence, Exercise, Exercise Prescription, Machine LearningAbstract
Introduction: artificial intelligence is revolutionizing exercise prescription in sports physiotherapy by offering more personalized and data-driven approaches. Through machine learning and deep learning algorithms, AI enables the analysis of complex variables such as biomechanics, physiological responses, and the patient's clinical history, dynamically adjusting exercise programs. This optimizes performance, prevents injuries, and enhances rehabilitation.
Methodology: a systematic review of studies on the use of AI in sports physiotherapy, based on articles published between 2015 and 2024, using specific inclusion criteria. The findings highlight the benefits of AI in personalizing exercise programs, emphasizing its capacity to improve adherence, load dosing, and injury prevention. However, clinical implementation of AI faces challenges such as external model validation, result interpretability, and the ethical management of sensitive data. Discussion: the review results show that AI is transforming exercise prescription in sports physiotherapy through a personalized and data-driven approach. AI algorithms, such as machine learning and deep learning, allow for the analysis of complex variables like biomechanics, physiological responses, and clinical history, dynamically adjusting exercise programs. Nevertheless, significant challenges remain for its clinical implementation, including external validation of models, interpretability of outcomes, and ethical concerns in handling sensitive data. Conclusion: AI holds tremendous potential to transform sports physiotherapy, but its integration into clinical practice requires overcoming technical and ethical challenges. Model validation, healthcare professional training, and equitable access to these technologies are essential aspects to ensure effective and safe implementation. Future research should address these challenges to maximize the benefits of AI in the field of exercise.
References
Topol EJ. High-performance medicine: the convergence of human and artificial intelligence. Nat Med. 2019;25(1):44-56.
Dey A, Salemt L. Artificial intelligence in physiotherapy: Transforming personalized exercise prescription. J Sports Sci. 2022;40(5):567-579.
Rathleff MS, Thorborg K. Load management and injury prevention: What is the role of artificial intelligence? Br J Sports Med. 2021;55(10):555-556.
Chen JH, Asch SM. Machine learning and prediction in medicine - Beyond the peak of inflated expectations. N Engl J Med. 2017;376(26):2507-2509.
Camacho DM, Collins KM, Powers RK, Costello JC. Next-generation machine learning for biological networks. Cell. 2018;173(7):1581-1592.
Esteva A, Robicquet A, Ramsundar B, Kuleshov V. A guide to deep learning in healthcare. Nat Med. 2019;25(1):24-29.
Tanaka R, Sato K, Matsumoto T. Artificial intelligence in exercise science: Current trends and future perspectives. Sports Med. 2020;50(12):2245-2258.
Claudino JG, Capanema DO, de Souza TV. Current approaches to monitor training load: The role of artificial intelligence. Front Physiol. 2019;10:843.
Peake JM, Kerr GK, Sullivan JP. Wearable technology and exercise prescription: Toward personalized activity recommendations. J Appl Physiol. 2018;124(3):692-702.
Yu KH, Beam AL, Kohane IS. Artificial intelligence in healthcare. Nat Biomed Eng. 2018;2(10):719-731.
García-de-Villa S, Casillas-Pérez D, Jiménez-Martín A, García-Domínguez JJ. Simultaneous exercise recognition and evaluation in prescribed routines: Approach to virtual coaches. Expert Syst Appl [Internet]. 2022;199(116990):116990. Disponible en:
http://dx.doi.org/10.1016/j.eswa.2022.116990
Spilz A, Munz M. Automatic assessment of Functional Movement Screening exercises with Deep Learning architectures [Internet]. arXiv [cs.LG]. 2022. Disponible en: http://arxiv.org/abs/2210.01209
Zhao Y, Liu P, Zhang X. Personalización del ejercicio en pacientes con lesiones deportivas mediante redes neuronales artificiales. J Sports Med Phys Fit. 2021;61(8):1205-1212.
Lee Y, Kim S. Optimización de la intensidad del ejercicio en rehabilitación postoperatoria mediante técnicas de machine learning y deep learning. J Med Eng Technol. 2021;45(2):88-95.
Thompson P, Rogers L, Harris J. Prevención de lesiones y recaídas mediante algoritmos de random forest para la dosificación de ejercicios. J Sports Sci Med. 2020;49(4):312-318.
Patel D, Chan A. Revisión de tecnologías wearables e inteligencia artificial en la personalización de la prescripción de ejercicio en fisioterapia. Med Devices. 2021;14:205-212.
Black H, Gray M, Carter S. Aplicación de algoritmos de clasificación en la rehabilitación de esguinces de tobillo utilizando máquinas de soporte vectorial. Phys Ther Sport. 2020;45(7):437-444.
Zhang C, Li Z, Zhou X, et al. Machine learning applications in injury prevention and rehabilitation in athletes: A systematic review. J Sci Med Sport. 2021;24(10):913-921.
Wong A, Bell C, Chiang Y, et al. Wearable sensors and artificial intelligence for exercise feedback in physical therapy: A pilot study. J Rehabil Res Dev. 2020;57(2):301-310.
Chung E, Ahn J, Kim K. AI-based personalized rehabilitation programs for post- surgical patients: Feasibility and clinical outcomes. BMC Health Serv Res. 2022;22(1):137.
Foster C, Hill M, Taylor S. AI and the future of sports medicine: A critical review. Sports Med. 2023;53(1):79-88.
Green T, Hall K, Wang J. Artificial intelligence in sports rehabilitation: Applications and challenges. Phys Med Rehabil Clin N Am. 2021;32(4):897-906.
Tanaka M, Matsumoto T, Higuchi T, et al. Effectiveness of AI-driven exercise programs in chronic musculoskeletal disorders: A meta-analysis. J Pain Res. 2022;15:1897-1907.
Yang X, Lu C, Li J, et al. Real-time monitoring of exercise techniques using wearable sensors and AI: Implications for injury prevention. Sports Health. 2021;13(4):304-310.
Park J, Shin H, Kim Y, et al. Deep learning models for exercise intervention in elderly individuals: A comprehensive review. Aging Clin Exp Res. 2023;35(2):223-235.
Published
Issue
Section
License
Copyright (c) 2025 Stalin Javier Caiza Lema, Paúl Adrián Arias Córdova, Angela Priscila Campos Moposita, Josselyn Gabriela Bonilla Ayala, Andrea Carolina Peñafiel Luna (Author)
This work is licensed under a Creative Commons Attribution 4.0 International License.
The article is distributed under the Creative Commons Attribution 4.0 License. Unless otherwise stated, associated published material is distributed under the same licence.
