Your company here — click to reach over 10,000 unique daily visitors

tend-bfit - Man Page

non-linear least-squares fitting of b-value curves


tend bfit [@file ...] [-i <nin>] -b <b1 b2 ...> [-w <w1 w2 ...>] [-imax <# iter>] [-eps <epsilon>] [-o <nout>]


Non-linear least-squares fitting of b-value curves. Axis 0 is replaced by three values: ampdecerr, based on a non-linear least-squares fit of amp*exp(-b*dec) to the range of DWI values along input axis 0, as a function of changing b values.


@file ...

response file(s) containing command-line arguments

-i <nin>

Input nrrd. List of DWIs from different b-values must be along axis 0; default: “-

-b <b1 b2 ...>

b values across axis 0 of input nrrd (2 or more doubles)

-w <w1 w2 ...>

weights for samples in non-linear fitting (2 or more doubles)

-imax <# iter>

max number of iterations to use in non-linear fitting, or, use 0 to do only initial linear fit (int); default: “10

-eps <epsilon>

epsilon convergence threshold for non-linear fitting (double); default: “1

-o <nout>

output tensor volume (string); default: “-

See Also


Referenced By


April 2021