Previous Topic

Next Topic

Book Contents

Book Index

Regression

The Regression auxiliary model performs a pixel-wise linear regression & correlation analysis.

Acquisition and Data Requirements

Image Data

Any dynamic volume data.

Volume Data

The pixel-wise linear regression uses the acquisition mid-times as the x-values. Therefore, the times must be set to appropriate values.

Note that the Regression model can be used to indirectly fit exponentials: To this end the exponentials are transformed into linear functions by selecting the Logarithm in the PREPROCESSING TOOLS section as illustrated below. When the linear regression is fitted to the transformed data, the slopes represent the exponents.

PXMOD Regression Volume Data

Model Configuration

Images of the following parameters can be generated

PXMOD Regression Model Parameters

slope

Slope of the regression line.

intercept

y-intercept of regression line.

corr

Pearson correlation coefficient for linear correlation (+/-1=complete correlation, 0=no correlation).

p

Significance level at which the null hypothesis of zero correlation is disproved (small value = significant correlation)