Bio-Inspired Controller on an FPGA Applied to Closed-Loop Diaphragmatic Stimulation.

Area of Science:

• Biomedical Engineering
• Neuroscience
• Computer Engineering

Background:

• Cervical spinal cord injury impairs brain-respiratory connections, necessitating ventilatory assistance.
• Current open-loop systems have limitations in adapting to changing patient needs.
• A closed-loop system offers a promising alternative for responsive respiratory support.

Purpose of the Study:

• To develop and simulate a bio-inspired, closed-loop assistive technology for real-time ventilation.
• To implement a spiking neural network (SNN) controller on a Field Programmable Gate Array (FPGA).
• To evaluate the system's performance using a computational model of rat breathing.

Main Methods:

• Designed a bio-inspired SNN controller mimicking the medullary respiratory network.
• Implemented the controller on a digital configurable FPGA with hardware constraints.
• Simulated the system in MATLAB using a computational rat breathing model.
• Tested both open-loop and closed-loop paradigms for real-time behavior.

Main Results:

• The bio-inspired controller demonstrated robustness comparable to classic controllers.
• FPGA implementation confirmed system functionality and real-time capabilities.
• Simulations showed the closed-loop system effectively monitors breathing and adjusts diaphragmatic stimulation.
• The model accurately reproduced experimental respiratory data.

Conclusions:

• The developed FPGA system represents a significant step towards adaptive, compact, low-power, implantable respiratory devices.
• The bio-inspired hardware design optimizes resource utilization and leverages neuromorphic computing.
• This real-time system is suitable for future in vivo applications and animal experiments.