You should only start training with the built-in cpsam model
I would like to know how to train a 12-channel model from scratch, e.g. HH3, panCK, CD20, CD45, FoxP3, CD3, CD4, CD8, CD14, CD31, CD56, CD66. I want to be able to separate touching cells that are incorrectly reported as one cell when so many cytoplasmic markers are combined into one value by a maximum summary.
I would like to know how to train a 12-channel model from scratch, e.g. HH3, panCK, CD20, CD45, FoxP3, CD3, CD4, CD8, CD14, CD31, CD56, CD66. I want to be able to separate touching cells that are incorrectly reported as one cell when so many cytoplasmic markers are combined into one value by a maximum summary.