Across our various projects, several students are doing things that require the same code, notably simulating a B0 distribution for a given susceptibility distribution and simulating MR signals in the presence of a given B0 distribution.
Matlab code has previously been developed for simulating a B0 distribution for a given susceptibility distribution: https://github.com/evaalonsoortiz/Fourier-based-field-estimation
This code is now being transcribed by @CharlesPageot into python: https://github.com/shimming-toolbox/susceptibility-to-fieldmap-fft
For simulating MR signals, we have again code in matlab:
https://github.com/evaalonsoortiz/B0_sim-mapping/tree/clean-rep
And python, as part of the shimming-toolbox package:
https://github.com/shimming-toolbox/shimming-toolbox/blob/master/shimmingtoolbox/simulate/numerical_model.py
I think we should use these codes where possible, and improve/expand them when needed, instead of re-implementing the same or similar, which is error-prone.
We should strive to make code modular, so that things can easily be re-used for different projects. This applies to @Danirid and @sriosq 's projects.
Across our various projects, several students are doing things that require the same code, notably simulating a B0 distribution for a given susceptibility distribution and simulating MR signals in the presence of a given B0 distribution.
Matlab code has previously been developed for simulating a B0 distribution for a given susceptibility distribution: https://github.com/evaalonsoortiz/Fourier-based-field-estimation
This code is now being transcribed by @CharlesPageot into python: https://github.com/shimming-toolbox/susceptibility-to-fieldmap-fft
For simulating MR signals, we have again code in matlab:
https://github.com/evaalonsoortiz/B0_sim-mapping/tree/clean-rep
And python, as part of the shimming-toolbox package:
https://github.com/shimming-toolbox/shimming-toolbox/blob/master/shimmingtoolbox/simulate/numerical_model.py
I think we should use these codes where possible, and improve/expand them when needed, instead of re-implementing the same or similar, which is error-prone.
We should strive to make code modular, so that things can easily be re-used for different projects. This applies to @Danirid and @sriosq 's projects.