lhsol 2.0.2 © 2002-2012 by Turku PET Centre
Fitting of full or reduced compartmental model to plasma and tissue
time-activity curves (TACs) to estimate the model parameters:
____ ____ ____ ____
| | K1 | | k3 | | k5 | | compartments in series (s)
| Cp |--->| C1 |--->| C2 |--->| C3 |
|____|<---|____|<---|____|<---|____|
k2 k4 k6
____ k3 ____
| |--->| | compartments in parallel (p)
____ | |<---| C2 |
| | K1 | | k4 |____|
| Cp |--->| C1 |
| |<---| | k5 ____
|____| k2 | |--->| |
| |<---| C3 |
|____| k6 |____|
Compartmental models are transformed into general linear least squares
functions (1, 2, 3), which are solved using Lawson-Hanson non-negative
least squares (NNLS) algorithm (4). Linear parameters are always >=0,
but compartmental model parameters may get negative estimates.
Note that rate constants and macroparameters are represented per volume
(measured by PET) including vascular volume.
Usage: lhsol <Model> <Plasma file> <Tissue file> <Fit time> <Result file>
, where Model is one of these strings:
k1 (for assuming k2=k3=k4=k5=k6=0)
k2 (for assuming k3=k4=k5=k6=0)
k3 (for assuming k4=k5=k6=0)
k4 (for assuming k5=k6=0)
k5s (for assuming k6=0 and compartments in series)
k6s (compartments in series)
k5p (for assuming k6=0 and compartments in parallel)
For model 'k6p' many of model parameters could not be solved.
Options:
-Vp=<ignored|fitted>
Vp is ignored (default) or fitted; note that plasma curve is assumed
to represent vascular blood curve, which is usually not true.
-w1 | -wf
Sample weights are set to 1 (-w1) or to frame lengths (-wf);
by default weights in tissue data file are used, if available.
-mid[=<y|N>]
Mid frame times are used (y) or not used (n, default) even if frame
start and end times are available.
-fit=<Filename>
Fitted regional TACs are written in DFT format.
-svg=<Filename>
Fitted and measured TACs are plotted in specified SVG file.
-par=<Filename>
Parameters of linear model are saved in result format.
-h or --help
Print this message and exit.
--version or --build
Print software build information and exit.
--silent
Program works silently, printing only error and warning messages.
--verbose
Program prints more information about what it is doing.
References:
1. Blomqvist G. On the construction of functional maps in positron emission
tomography. J Cereb Blood Flow Metab 1984;4:629-632.
2. Gjedde A, Wong DF. Modeling neuroreceptor binding of radioligands
in vivo. In: Quantitative imaging: neuroreceptors, neurotransmitters,
and enzymes. (Eds. Frost JJ, Wagner HM Jr). Raven Press, 1990, 51-79.
3. Oikonen V. Multilinear solution for 4-compartment model:
I. Tissue compartments in series.
http://www.turkupetcentre.net/reports/tpcmod0023.pdf
4. Lawson CL & Hanson RJ. Solving least squares problems.
Prentice-Hall, 1974.
See also:
dftcbv,
patlak,
logan,
lhsolki,
lhsoldv,
fitk2,
fitk3,
fitk4
Keywords: DFT, modelling, compartmental model, multilinear fit, NNLS
This program comes with ABSOLUTELY NO WARRANTY. This is free software, and
you are welcome to redistribute it under GNU General Public License.