Back to the Top
The following message was posted to: PharmPK
Can someone point me to a reference or algorithm on curve stripping?
More
specifically, I need to know how to statistically determine the time
point
where one phase breaks from another.
Thanks,
Cory Langston
Cory Langston, DVM, PhD, DACVCP
College of Veterinary Medicine
Box 6100 (Spring Street for courier)
Miss. State, MS 39762-6100
phone 662-325-1265
fax 662-325-4011
email langston.-a-.cvm.msstate.edu
Back to the Top
The following message was posted to: PharmPK
Cory,
Our pharmacokinetics software, PK Solutions, provides the most flexible
approach to curve stripping. You are invited to take a look at the demo
at www.SummitPK.com. Unlike compartmental curve fitting, which follows
certain standard methods to narrow down (supposedly) the best fit
model, curve stripping puts your judgement in the driver's seat. As you
can see with PK Solutions, we provide an immediate feed back of the
graphical over lay of the regression line over your data points as well
as the correlation coefficient, and standard error of your selected
linear region(s). With these parameters in view, you can optimize your
selection of the coefficients describing the various phases during the
curve stripping process.
Best wishes,
David
David S. Farrier, Ph.D. Phone: 970-249-1389
Summit Research Services Fax:: 970-249-1360
68911 Open Field Dr. Email: DFarrier.aaa.SummitPK.com
Montrose, CO 81401 Web: http://www.SummitPK.com
Back to the Top
The following message was posted to: PharmPK
Dear Cory,
There are some old references on curve stripping, including code in
BASIC
for automated performance. The curve-stripping method is a nice tool to
obtain initial estimates for pharmacokinetic parameters. However, by
now we
have numerous programs for curve-fitting, providing 'the best estimates'
based on statistical theory (and many assumptions!).
In my view, the estimates from a stripping method should not regarded as
reliable parameter values.
> I need to know how to statistically determine the time point
> where one phase breaks from another.
This time point does not exist. The only reliable way to perform curve
stripping is to test systematically all combinations of adjacent data
points
used for the two or three phases, and to select the combination
resulting in
the lowest sum of squared weighted residuals (measured value minus the
calculated value using the complete polyexponential equation).
In my view, making such a program makes no sense anymore. E.g. using the
Excel Solver you can solve simple fitting problems within minutes.
Best regards,
Hans Proost
Johannes H. Proost
Dept. of Pharmacokinetics and Drug Delivery
University Centre for Pharmacy
Antonius Deusinglaan 1
9713 AV Groningen, The Netherlands
tel. 31-50 363 3292
fax 31-50 363 3247
Email: j.h.proost.-at-.farm.rug.nl
Back to the Top
Dear Cory:
You have put your finger on a key problem of curve stripping.
It was OK in the 1960's, when computing power was very small. Most
people do not do this any more, as there are much better means to do
this, for example, any population modeling software. The nonparametric
approach in the USC*PACK programs if more consistent (the more subjects
you study, the closer the results are to the true ones), and efficient
than any others we have compared them with, such as IT2B and NONMEM.
You might consider using these really much better methods. Further,
what weighting scheme do you propose for your data? How will
conventional curve stripping give weight to each data point according
to a measure of its credibility?
Our web site iswww.lapk.org. You can go there for more
information.
Very best holiday wishes,
Roger Jelliffe
Roger W. Jelliffe, M.D. Professor of Medicine,
Division of Geriatric Medicine,
Laboratory of Applied Pharmacokinetics,
USC Keck School of Medicine
2250 Alcazar St, Los Angeles CA 90033, USA
Phone (323)442-1300, fax (323)442-1302, email= jelliffe.-at-.usc.edu
Our web site=http://www.lapk.org
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)