Content-type: text/html
0. No problems.
6. Bad data replaced by substitution of zero, see modelfit(1).
The matrix must be R by (N+M), where N+M is the number of measurements and R is the number of parameters of the reconstruction. The matrix file contains binary double-precision floats. The matrix elements are stored row by row.
If the first command line argument is - the program reads data from the standard input.
$ # Create the DT fitting matrix
$ dtmx -schemefile A.scheme > A_DTMX.Bdouble
$ # Run it over each voxel
$ linrecon SubjectA.Bfloat A.scheme A_DTMX.Bdouble -log > DiffTensorA.Bdouble
The above is equivalent to:
$ dtfit SubjectA.Bfloat A.scheme > DiffTensorA.Bdouble
To save a matrix from matlab in the right format, follow this example:
>> % The matrix has size 755x54. 755 reconstruction parameters and 54
>> % normalized measurements.
>> size(rbfkernel)
ans =
755 54
>> fid=fopen('/tmp/Q_BallBMX7_Matrix.Bdouble', 'w', 'b');
>> fwrite(fid, rbfkernel', 'double');
>> fclose(fid);
Then run
$ linrecon SubjectA.Bfloat A.scheme /tmp/Q_BallBMX7_Matrix.Bdouble -normalize > /tmp/QBALL_A.Bdouble
The matrix Q_BallBMX7_Matrix.Bdouble is for Q-Ball reconstruction, which uses the normalized measurements, hence the -normalize flag.