MAISI mask autoencoder

Hi,

Thanks for your great work with MAISI. The mask_generation_autoencoder has 8 input channels. What is the kind of data that is required as input?
Would you have an example on how to use the mask autoencoder alone, without the diffusion part based on the provided data description for mask_pseudo_label*.nii.gz ?

"mask_generation_autoencoder_def": {
    "_target_": "monai.apps.generation.maisi.networks.autoencoderkl_maisi.AutoencoderKlMaisi",
    "spatial_dims": "@spatial_dims",
    "in_channels": 8,
    "out_channels": 125,

Thanks,
Christian

Start with the MAISI mask_generation_autoencoder configuration and the provided mask_pseudo_label*.nii.gz data description. Determine the expected eight-channel input and document a standalone mask autoencoder example; done when the input format and usage are clear without relying on the diffusion workflow.