AI-Enabled Precision Neurorehabilitation in Sports: Integratings Neuroplasticity, Brain–Muscle Connectivity, Wearable Biomarkers, and Personalized Return-to-Play Decision-Making

Authors

  • Muhammad Mustafa Chief Consultant Rehabilitation, Shanghai Elite Sports Rehabilitation Center, China Author
  • Dr. Kinza Afzal Senior Lecturer, Hamdard University, Islamabad Campus Author
  • Sorath Sadia PT, College of Physiotherapy, JPMC Author
  • Dr. Safia Arfeen PUMHSW, Shaheed Benazirabad Author
  • Dr. Salman Naeem Head of Anesthesia, ICU and Pain Management, Mian Muhammad Trust Hospital, Faisalabad Author
  • Zulkaif Asghar PT, University of Sargodha Author
  • Muhammad Imran Iqra University, Chak Shahzad Campus, Islamabad Author

DOI:

https://doi.org/10.71317/jgst.2.9(s4).2026.759

Keywords:

artificial intelligence, neurorehabilitation, neuroplasticity, brain–muscle connectivity, wearable biomarkers, return-to-play, sports medicine

Abstract

Background: Traditional return-to-play (RTP) protocols in sports medicine heavily rely on subjective symptom reporting and isolated musculoskeletal assessments, often overlooking central neuroplastic adaptations and brain–muscle connectivity following injury. Objective: This study aimed to evaluate the efficacy of an AI-enabled precision neurorehabilitation framework that integrates neuroplasticity metrics, brain–muscle connectivity, and wearable biomarkers to personalize and optimize RTP decision-making. Method: A prospective longitudinal cohort study was conducted with 142 elite athletes (mean age = 24.3 ± 4.1 years) recovering from sport-related concussions or anterior cruciate ligament (ACL) reconstructions at the Pakistan Sports Complex in Islamabad, Pakistan. Multimodal wearable sensors (EEG, sEMG, IMUs, and ECG) continuously captured neuroelectric, kinematic, and autonomic biomarkers throughout rehabilitation. A Long Short-Term Memory (LSTM) neural network processed the high-dimensional time-series data to generate a continuous "Neuro-Readiness Score" guiding clinical RTP decisions. Results: The AI predictive model demonstrated high accuracy for RTP timeline prediction (89.4%) and six-month re-injury risk (AUC = 0.88). Longitudinal analysis revealed significant restoration of EEG–EMG coherence in the beta band and the normalization of cortical load by Week 12. Crucially, the AI framework identified a subset of athletes (n = 18) with concealed neurological deficits despite being clinically asymptomatic, prompting a targeted delay in their RTP and mitigating re-injury risk. Conclusion: AI-enabled precision neurorehabilitation successfully transitions sports medicine from reactive, time-based protocols to predictive, neuro-centric paradigms. By continuously monitoring brain–muscle connectivity and autonomic stress, this framework provides objective, individualized RTP metrics, significantly enhancing athlete safety and proving highly viable for implementation in emerging elite sports environments.

Downloads

Published

2026-09-23

How to Cite

Mustafa, M., Afzal, K., Sadia, S., Arfeen, S., Naeem, S., Asghar, Z., & Imran, M. (2026). AI-Enabled Precision Neurorehabilitation in Sports: Integratings Neuroplasticity, Brain–Muscle Connectivity, Wearable Biomarkers, and Personalized Return-to-Play Decision-Making. Journal of Global Social Transformation, 2(9.4), 185-192. https://doi.org/10.71317/jgst.2.9(s4).2026.759