Abstract
Linear parametric neurotransmitter PET (lp-ntPET) is a novel kinetic model that estimates the temporal characteristics of a transient neurotransmitter component in PET data. To preserve computational simplicity in estimation, the parameters of the nonlinear term that describe this transient signal are discretized, and only a limited set of values for each parameter are allowed. Thus, linear estimation can be performed. Linear estimation is implemented using predefined basis functions that incorporate the discretized parameters. The implementation of the model using discretized parameters poses unique challenges for significance testing. Significance testing employs model comparison metrics to determine the significance of the improvement of the fit accomplished by including a basis function, i.e. it determines the presence of a transient signal in the PET data. A false positive occurs when the bases overfit data that do not contain a transient component. The number of parameters in a model, p, is necessary to determine the degrees of freedom in the model. In turn, p is crucial for the calculation of model selection metrics and controlling the false positive rate (FPR). In this work, we first explore the effect of parameter discretization on FPR by fitting simulated null data with varying numbers of bases. We demonstrate the dependence of FPR on number of bases. Then, we propose a correction to the number of parameters in the model, peff , which adapts to the number of bases used. Implementing model selection with peff maintains a stable FPR independent of number of bases.
Highlights
P OSITRON emission tomography (PET) makes it possible to image molecular targets with high specificity
false positive rate (FPR) increased at a saturable rate with number of bases for each model comparison metric (Fig. 5)
FPR was uniformly below 5% for all model comparison metrics when calculated with p f ull = 7
Summary
P OSITRON emission tomography (PET) makes it possible to image molecular targets with high specificity. The linearparametric neurotransmitter model (lp-ntPET) estimates the timing of transient neurotransmitter (NT) release occurring within a single scan [1]–[7]. Because of its linear form, lp-ntPET can be used to estimate parameters that describe the tracer and NT components on the voxel level with high computational efficiency— thousands of voxels can be fitted, in just minutes. Parameters in the NT component that describe the timing of the transient signal are discretized. Each discrete combination of possible timing parameters forms one basis function. All combinations together form a library of ‘bases’ that represents all candidate timing profiles of the transient signal. The combination of linear parameters and basis function that produces the best fit is retained as the set of optimal parameters
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
