Does it supports nested region segmentation like nnUNetv2?

Hi,
I am trying to use this model for nested region segmentation like nnUNetv2 like the following

"labels": {
        "background": 0,
        "subject": [1,6,7,9],
        "deflated_device": 2,
        "inflated_device": 3,
        "vessel": [4,6,8,9],
        "wire": [5,7,8,9]
    }

As this code is based on nnunet, I am curious if we can do it with TransUNet. If yes, how to do it?


