Related Experiment Video
Updated: Jun 28, 2026
04:37
A Passive Ankle Dorsiflexion Testing System for an In Vivo Model of Overuse-induced Tendinopathy
Published on: March 1, 2024
A physics-informed deep learning framework for estimating muscle activation patterns following Achilles tendon repair
Diwei Chen1, Dong Sun1,2,3, Yufei Li4
1Faculty of Sports Science, Ningbo University, Ningbo, People's Republic of China.
Journal of Applied Physiology (Bethesda, Md. : 1985)
|June 26, 2026
Summary
A new physics-informed Transformer Neuromusculoskeletal Model (NMM) accurately estimates soleus (SOL) muscle activation for Achilles tendon rupture (ATR) recovery. This model offers a practical clinical tool for assessing deep muscle recruitment and co-activation patterns.
Area of Science:
- Biomechanics
- Neuromuscular modeling
- Rehabilitation engineering
Background:
- Accurate soleus (SOL) muscle activation assessment is crucial for Achilles tendon rupture (ATR) recovery.
- Direct measurement of deep muscle activation presents a significant clinical challenge.
Purpose of the Study:
- To develop and validate a physics-informed Transformer Neuromusculoskeletal Model (NMM) for estimating SOL activation during locomotion.
- To assess muscle co-activation patterns in ATR patients during recovery.
Main Methods:
- A Transformer-based Neuromusculoskeletal Model (NMM) integrating forward dynamic constraints was developed.
- The NMM utilized joint kinematics and minimal surface electromyography data.
- Performance was benchmarked against conventional methods (SO, SYNX) and applied to 40 participants (20 healthy, 20 ATR patients).
Main Results:
- The NMM demonstrated superior spectral integrity and signal energy compared to SO and SYNX (p < 0.05).
- ATR patients exhibited a phase-dependent shift in SOL recruitment: reduced early-stance and elevated terminal-stance activation (p < 0.01).
- Co-activation analysis revealed altered early- and late-stance muscle recruitment patterns (p < 0.05) and a shift towards stability-prioritizing strategies.
Conclusions:
- The NMM framework effectively characterizes soleus muscle co-activation patterns during Achilles tendon rupture recovery.
- This model offers a practical, streamlined clinical tool for gait monitoring, reducing sensor requirements while maintaining diagnostic depth for deep muscle recruitment.
