The road toward a physiological control of artificial respiration: the role of bio-inspired neuronal networks.

The transition from mechanical to physiological ventilation is a delicate step during the recovery from ECMO, in particular following severe respiratory failures. Since there is controversy on the optimal degree of mechanical ventilation support, the maintenance of physiological ventilation can be crucial to determine the balance between lung rest and lung recovery. We believe that the development of closed-loop control systems for mechanical ventilation, designed to maintain or restore physiological respiratory activity in patients supported by extracorporeal membrane oxygenation (ECMO) could contribute to achieve this goal. In our vision, the core of such a system could be a biologically inspired computational model of the respiratory neural control center, capable of simulating the respiratory rhythm required to efficiently eliminate CO₂ from the body. The outputs of the modeled respiratory rhythm (e.g., rate and pattern) would represent the patient's needs that should be ideally maintained to ensure proper CO₂ clearance. The use of a simulated respiratory rhythm to dynamically control a mechanical ventilator integrated with ECMO would ensure that ventilatory support is adjusted in real time to meet the physiological demands indicated by inputs delivered by external sensors. One of the key advantages of this system would be its use during weaning from ECMO. By simulating a target respiratory rhythm and gradually transferring the workload from ECMO to mechanical ventilation, the system could allow for a smoother and safer transition to spontaneous or assisted breathing.