Sorry for the delayed response. I was traveling when the initial message was sent & this slipped past me. The easiest way to do this is to use the --mix option to supply the ICA mixing matrix from a previous run of tedana. There will be a few things that are unnecessarily re-calculated, but the most computationally intensive step (PCA & ICA) will be skipped and you should get identical results through the decision tree step. If you want to compare decision tree results, you can use dtm_tools.
FWIW, my long-term to do list includes adding this functionality to ica_reclassify to avoid all repeated calculations. I’m not sure when I’ll get into it.