Non-invasive estimation of input function in ultra-rapid CBF/CMRO2 measurement

本研究的目的是建立一种无创的快速PET测量脑血流量（CBF）和氧代谢（cro_2）的方法。该方法通过引入模型函数来表示输入函数的形状和尺度，然后分别计算水和氧的组织响应曲线，即得到模型组织曲线。将模型组织曲线拟合到实测的多个脑组织曲线上，得到模型函数的参数，对输入函数进行重构。在测量数据中，采用无创方法测定CBF和cmo_2的准确度为13%。仿真结果表明，该方法的精度在10%左右。这些结果支持了该方法可用于CBF和cmo_2 PET检查。

The Purpose of the present study is to develop a non-invasive method for obtaining cerebral blood flow (CBF) and oxygen metabolism (CMRO_2) in rapid PET measurement. In the present approach, a model function was introduced to express a shape and scale of input function, and then tissue response curves for water and oxygen, respectively, was calculated, i.e., model tissue curves were obtained. The model tissue curves were fitted to the measured multiple brain tissue curves and obtained parameters for the model function reconstructed the input functions. In the measured data, the accuracy of CBF and CMRO_2 were 13% using the non-invasive procedure. Also, simulation showed that the accuracy for the present method is around 10%. All these findings supported that the present method would of use for CBF and CMRO_2 PET examination.