Shiji Li

About

Shiji Li is a leading researcher at the intersection of biomedical signal processing, rehabilitation robotics, and human-machine interaction. Their work focuses on decoding complex neuromuscular signals to enable intuitive control of assistive technologies, particularly for lower limb applications. Li’s most notable contribution is the development of a dual-branch deep learning framework that simultaneously classifies lower limb motions and estimates continuous joint angles from surface electromyography (sEMG) signals. This framework, enhanced with transfer learning, addresses a critical challenge in wearable robotics: achieving both accurate motion recognition and real-time joint angle prediction from the same neuromuscular data. By integrating dual-branch architectures with transfer learning, Li’s approach significantly improves the robustness and generalizability of sEMG-based control systems, reducing the need for extensive retraining across different users or conditions. This work has immediate implications for exoskeleton control, prosthetic limb design, and rehabilitation therapy, where seamless, naturalistic movement assistance is paramount. With their 2025 paper already garnering early citations, Li is establishing a strong foundation for future innovations in adaptive, human-centered robotic systems that respond to the user’s own biological signals.

Research Focus

Key Achievements

Top Papers

1. 1
   <scp>EMG</scp>‐Based Dual‐Branch Deep Learning Framework With Transfer Learning for Lower Limb Motion Classification and Joint Angle Estimation

   4 citations · 2025

Key Collaborators

Contact & Links