Movatterモバイル変換

gastempt: Fittinggastric emptying curves

Dieter Menne
Menne Biomed Consulting Tübingen, Germany
https://www.menne-biomed.de
dieter.menne@menne-biomed.de

A package and aShiny webapplication to create simulated gastric emptying data, and to analyzegastric emptying from clinical studies using a population fit with R andpackagenlme. In addition, Bayesian fits withStan to handle critical cases areimplemented.

Part of the work has been supported by section GI MRT, Klinik fürGastroenterologie und Hepatologie, Universitätsspital Zürich; thanks toWerner Schwizer and Andreas Steingötter for their contributions.

Download

The package is available fromCRAN and github(source,documentation). Toinstall, use:

devtools::install_github("dmenne/gastempt")

Compilation of the Stan models needs several minutes.

Shiny online interface

The web interface can be installed on your computer, or run asweb app.

Twomodels are implemented in the web interface

• linexp, vol = v0 * (1 + kappa * t / tempt) * exp(-t / tempt):Recommendedfor gastric emptying curves with an initial volume overshoot fromsecretion. With parameter kappa > 1, there is a maximum after t=0.When all emptying curves start with a steep drop, this model can bedifficult to fit.
• powexp, vol = v0 * exp(-(t / tempt) ^ beta): The powerexponential function introduced by Elashof et. al. to fit scintigraphicemptying data; this type of data does not have an initial overshoot bydesign. Compared to thelinexp model, fittingpowexp is more reliable and rarely fails to converge in thepresence of noise and outliers. The power exponential can be useful withMRI data when there is an unusual late phase in emptying.

Methods with variants

• Population fits based on functionnlme in package Rnlme.
• Stan-based fits, bothwithoutandwithcovariance estimation. Thanks to priors, fitting with Bayesian methodsalso works for single records, even if stability strongly improves withmore data sets available; seestan_gastempt.Some details can be found inmy blog. Therationale for using Stan to fit non-linear curves is discussedhere for¹³C breath test data, but is equally valid for gastricemptying data.

Data entry:

• Data can be entered directly from the clipboard copied from Excel,or can be simulated using a Shiny app.
• Several preset simulations are provided in the Shiny app, withdifferent amounts of noise and outliers
• Robustness of models can be tested by manipulating noise quality andbetween-subject variance.
• Fits are displayed with data.
• The coefficients of the analysis including half-emptying timet50 and the slope at timet50 can be downloaded in.csv format.

Example

Program with simulated data (needs about 40 seconds till plotshows):

library(gastempt)dd = simulate_gastempt(n_records = 6, seed = 471)d = dd$dataret = stan_gastempt(d)print(ret$coef)print(ret$plot)

Installing Docker

• For Windows 10, you can get the installer from theDockerstore. For installation details, seehere.
  
• Linux users know how to install Docker anyway.

Installinggastempt

• From the command line, enter the following to start thecontainer

docker run --name gastempt  --restart unless-stopped -p 3838:3838 -d dmenne/gastempt

• The first startup needs some time because 1 GB has to be downloaded.Subsequent startups require only a few seconds.
• Connect to the app with your browser vialocalhost:3838.