sonukumari1990/mri-preprocessing-brain-tumour ? reverse-engineered prompt
Reverse engineered prompt
Build me a Python research project that reproduces this MRI preprocessing study from the saved run results.
I want the repo to let me install the requirements, then generate all the paper tables and figures from the existing results files in a couple of minutes on a normal laptop, with no GPU and no need to download imaging data. The project should also include the training and preprocessing scripts used for the study, plus clear docs for how to rerun everything from the original images if needed.
Make sure the code can handle the factorial setup of preprocessing choices, the three CNN models, the two datasets, and the three prediction tasks, and that it produces the same summaries, tables, and figures described in the README. Also include the small checks and mechanism analyses around N4 bias field correction and the task versus site comparisons.
Please organize it so the main analysis scripts are easy to run directly, and if you need to look up current docs or library behavior online, go ahead.
Are you gonna build this?
make sure you review the code using arcumet