Identification of physiological lung parameters using the forced oscillation technique

Medical doctors only have access to limited information about the patient's respiratory system when ventilating a patient. Only a first order model is used for the patient's respiratory system for patient ventilation applications where a patient's breathing is assisted or taken over by a machine. We propose the use of frequency domain identification along with the Forced Oscillation Technique javascript:void(0);(FOT, also referred to as oscillometry techniques) to provide more physiological parameters linked to the patient's lung condition. A constant phase model is used which is known to deliver a wider set of parameters which are physiologically interpretable and can now be parametrically identified for patients being ventilated. First results on a healthy subject are provided. The design of the excitations used is also detailed. It requires the spectral separation of the controlled excitation and the patient's estimated breathing. The latter is indeed modelled as a disturbance in the low frequency range. Additionally, the SNR is typically very low, making the identification challenging