本研究は、システム同定によって患者の肺特性である肺内圧力を推定し、人工呼吸器設定にフィードバックする手法を開発している。本論文では、患者の肺特性をファジィモデルにおいて表現しており、それに係るファジィ変数を最適化する手法を提案している。

The purpose of this study is to develop a method to estimate human pulmonary characteristic and to set a ventilation condition of the artificial respirator.
Systematic technique for setting the ventilation condition of the artificial respirator is not develop.
The artificial respirator is set based on the experience and the intuition of the doctor. Our study attempts to solve this problem. Firstly, this study estimates a pulmonary characteristic using a measurement data of the patient. Then, the artificial respirator is set using estimated pulmonary characteristic.
These processes are carried out using a fuzzy logic.
By using same logic for all processes, we will expect the information sharing of each process.

The characteristic of pulmonary is comprised of the pulmonary internal pressure and the pulmonary volume. However, we do not understand the pulmonary internal pressure, because it cannot measure the pulmonary internal pressure by the respirator. Then, the　characteristic of pulmonary is different, and the setting of the respirator is decided by the　experience and the perception of the doctor now.
Therefore, the suitable setting of ventilation conditions is needed for every patient. The static P-V curve expresses the characteristic of human lungs, and that is needed for the parameter setting of the respirator. Purpose of this study is to develop a method to estimate the static P-V curve and the pulmonary elastance of the patient. In our previous work, we proposed technique that separated parameters estimation procedure of the pulmonary elastance and parameters estimation procedure of the other terms.
In this paper, a recursive estimation method of the pulmonary elastance model's parameters is addressed to improve the accuracy of estimation model. Some examples by using practical clinic data are given to show the effectiveness of addressed method.