Regression Mysteries
Dieter Menne 15 Juli 2017
In the paper by Hoad et al. (2015), two different methods to measure gastric content volume (GCV) were compared, by MRI and by gamma scintigraphy. A regression line was computed which is the best prediction of GCV(MRI) given GCV(Scintigraphy). The s...
Box-Cox Transformations for linexp fits
January 15, 2020, 7:09 pmThe coefficients of linexp fits to gastric emptying curves are generally used in mixed-models to analyze group and meal difference. The distribution of the extracted parameters is often highly skewed, so before using linear models or estimates of reference ranges with functions in package referenceInterval or Hmisc, a transformation might be required.
From a large corpus of gastric emptying curves analyzed with nonlinear fits of the linexp function, the following recommendations for transformation were obtained using the boxcox and the logtrans function in package MASS:
Gastric emptying can be recorded by scintigraphy or by MRI techniques. Scintigraphy returns times series data of the meal volume, MRI gives the total content of the stomach, which includes meal and secretion. To quantify secretion and emptying in clinical studies, non-linear curves are fitted to the time series, which can give strange results. When the ballot fails, the bazaar comes to the rescue.