I’ve updated the scripts for graphing our stability metrics using R rather than Matlab (thanks, ChatGPT!). Let me know what you think. One note, the change in scanning protocol to ensure the phantom was really in isocenter for the 64-channel coil tests is really evident in the graph below, particularly for bringing down the measures of ghosting artifact. I think we’ve really got it under control now, but it highlights the importance of isocentering your subjects.
