The Modeling And Study Of Human Ocular Blood Flow Autoregulation Based On LS

At present, the ocular blood autoregulation dysfunction have been proved to be one of the independent risk factors of many ocular diseases. There is no direct cause of the autoregulation system damaged widely accepted, no method can directly protect autoregulation system be impaired.And the studies of ocular blood flow autoregulation did not delve into the physiological mechanism of ocular blood flow system to ocular blood flow autoregulation system modeling research.The study of ocular blood flow autoregulation system can help human understand the supply mechanism of ocular blood flow, and it can provide the basis for the diagnosis, treatment and prevention of disease. The experimental data of this study collected the signals of ocular blood flow (OBF) by using LSFG,and at the same time collected the signals of ocular blood pressure (OBP) by using Finometer under the experimental conditions of the thigh cuff. Ocular blood pressure and ocular blood flow were concentrated in the low frequency energy cycle of non-stationary random signal, and in the process of collecting were vulnerability to instrument, respiration, action, usually with strong noise. In this paper, wavelet filtering and other denoising methods are adopted to remove baseline drift, power frequency interference, myoelectricity interference noise, etc.Ocular blood flow autoregulation mechanism define the ocular blood flow remains relatively constant in certain range. The present study of the latest direction of ocular blood flow autoregulation mechanism is establishing the linear or nonlinear model of the system. In this paper, the aim is establishing the model of the OBF autoregulation system by using the system identification technology under the experimental conditions of the thigh cuff. The technology of model parameter identification selected LSM algorithm. LSM is one of the most widely classical data processing method which is used to solve some practical problems. Because during the frequency range of 0.05-0.3Hz, the relationship between the OBF and OBP shows more performance for linear relationship. So in this paper, the linear ARX model system model was established, then transform it to the model of system transfer function parameters.At last, analysis the stability of the model and verify the validity of the model.In order to fully understand the OBF autoregulation system of linear and nonlinear characteristics, in this paper also attempts to establish a nonlinear model of the system. Given the nature of the poor repeatability of nonlinear model, this paper puts forward combining the linear and nonlinear system model is established the Hammerstain model. The predict data of this model and actual data fitting rate is 86.81%, and the fitting rate of ARX model is 83.92%.The repeatability of Hammerstain model is better than ARX model. When using another set of samples to verify the two models, the fitting rate of Hammerstain model is 82.07%, and the fitting rate of ARX model is 63.7%.