Effects of noise and bias in
signal and different parameters in the kinetic model affects the
reproducibility of the estimation of the apparent rate of conversion of
hyperpolarized (HP) pyruvate into HP lactate (kpl). The purpose of this
study is to investigate the effect of signal to noise ratio for a
pharmacokinetic model with two chemical and two physical compartments in
estimating kpl. We examine the kpl estimated using the
kinetic model by the simulated HP data with a variety of SNRs (10~50) by 95%
confidence interval, mean and standard deviation. The results demonstrate that
SNR affects the reproducibility of the estimation of kpl by kinetic
analysis and the reproducibility of kpl estimated decreases quickly below an
SNR threshold of ~25.
