Dear Stefano,
I'm still intrigued by this problem and have tried multiple 'variations on the theme' in attempts to get NiftyRec software running.
I've downloaded NiftyRec-1.6.9 and been playing around with some debugging - on the basis of various error messages generated when trying to run NiftyRec 1.6.8 - and have noticed an issue which could greatly impact on my efforts:
Its to do with et_demo_fbp.m where the code is as follows:
<code>
% ET_DEMO_FBP
% NiftyRec Demo: FBP SPECT reconstruction
%
%See also
% ET_DEMO_MLEM, ET_DEMO_OSEM, ET_DEMO_MAPEM_MRF
%
%
%Stefano Pedemonte
%Copyright 2009-2013 CMIC-UCL
%Gower Street, London, UK
%% Parameters
N = 128;
N_projections = 120;
cameras = linspace(0, 2*pi, N_cameras);
psf = ones(5,5,N);
N_counts = 50e6;
iter_mlem = 30;
GPU = 1;
phantom_type = 0; % 0 for 'brain FDG PET'; 1 for sphere in uniform background; 2 for spiral
%% Initialise plot
figure(1); subplot(2,3,1);
</code>
I note that N_cameras insn't defined for:
cameras = linspace(0, 2*pi, N_cameras);
The code reads like N_cameras is meant to be N_projections (from the line above).
Does this indicate that the software is buggy and not yet suitable for deployment in the public domain?
Should I continue to try to get this software running?
Yours sincerely,
Kieran
Kieran Maher, Msc, PhD
Senior Physicist
Medical Imaging Dept
St Vincent's Hospital Melbourne
Australia
-=-
|