Electrical Stimulation System for Respiratory Assistance Based on Iterative Learning Control

Respiratory dysfunction significantly affects the quality of life for patients with chronic obstructive pulmonary disease, spinal cord injury, and stroke. Although neuromuscular electrical stimulation (NMES) has shown potential in respiratory rehabilitation, existing systems primarily rely on open-loop control and lack real-time adaptability. This study proposes a closed-loop multi-electrode respiratory neuromuscular electrical stimulation (R-NMES) system that integrates a respiratory flow sensor and an iterative learning control strategy to dynamically adjust both the period and intensity of the stimulation signal. By continuously optimizing R-NMES parameters in real time, the system enables the induced muscle contractions to gradually approach the natural respiratory rhythm. Experimental results demonstrate that R-NMES assistance significantly increases mean airflow velocity and minute ventilation, indicating that the system effectively enhances respiratory efficiency and offers a new approach for personalized and adaptive respiratory rehabilitation.