Content-type: text/html Manpage of datastats

datastats

Section: User Commands (1)
Index Return to Main Contents

NAME

datastats - Computes simple statistics of each diffusion weighted measurement within a region of interest. Useful for estimating the noise level.

SYNOPSIS

datastats [-schemefile <filename> -bgmask <filename>]

DESCRIPTION

Reads a data file from the acquisition scheme specified in the schemefile. The ROI can be specified in a mask file or by thresholding the b=0 image using -bgthresh. The program computes the mean, mean squared, variance and standard deviation of each image in the foreground region specified by the mask file.

EXAMPLES

Create a mask for data file SubjectA.Bfloat:

mask -inputfile SubjectA.Bfloat -schemefile A.scheme -bgthresh 200 -outputdatatype short > SubjectA_M200.Bshort

Compute the statistics:

datastats -inputfile SubjectA.Bfloat -schemefile A.scheme -bgmask SubjectA_M200.Bshort

Foreground voxel count: 3248

Component E(S) E(S^2) Var(S) Std(S)

1          1.388036E02    2.061075E04    1.344318E03    3.666494E01
2          1.376616E02    2.016732E04    1.216598E03    3.487977E01
3          1.378830E02    2.030152E04    1.289794E03    3.591370E01
4          1.391675E02    2.067510E04    1.307506E03    3.615945E01
5          1.381736E02    2.036596E04    1.274002E03    3.569317E01
:
:

Useful for estimating the noise level (as required for restore or mbalign) or signal to noise. An estimate of the noise level sigma (standard deviation of each component of the complex noise on the signal) is sqrt(E(S^2)/2) from an ROI entirely in background. Eg, draw an ROI in background on the FA image from the data set using MRIcro. (The mask volume should be the same size as one component of the full DWI data set.) Save the mask as an analyze image ROI->Export ROI as Analyze Image; Save [Sun]. MRIcro saves the mask as type char, so to measure the statistics in the region run:

datastats -inputfile SubjectA.Bfloat -schemefile A.scheme -bgmask A_BG.img -maskdatatype char

If the ROI is truly in background, you should see little variation among the different diffusion weighted images, so can choose any to compute sigma.

OPTIONS

-inputfile <file>: Diffusion-weighted MRI data file in voxel order.
-schemefile <file>: Schemefile.
-bgthresh <integer>: The threshold on the average b=0 measurement.
-inputdatatype <data type of input>: Specifies the data type of the input file. The data type can be any of the following strings: "char", "short", "int", "long", "float" or "double". See camino(1).
-outputdatatype <data type of output>: Specifies the data type of the output data. The data type can be any of the following strings: "char", "short", "int", "long", "float" or "double". See camino(1).
-bgmask <Mask file>: Mask file name. The mask file contains zero in background voxels and non-zero in foreground. The mask must be big endian; the default data type is 16-bit shorts, but can be changed using the -maskdatatype option.
-maskdatatype <char|short|int|long|float|double>: Specifies the type of the mask file; must be big-endian byte ordering.

AUTHORS

Daniel Alexander <camino@cs.ucl.ac.uk>

BUGS

No known bugs.

This document was created by man2html, using the manual pages.
Time: 02:07:11 GMT, December 04, 2017

Datastats