Hi, I am new to MONAI
I have 5 manually extracted brain features volumetric maps (thickness, sulc, curv, etc) and ROI target label mask with size (241, 336, 283) for a number of subjects. How can I stack all feature in inputs channels? and what classification model best fits for label detection?
Assuming each of your features has a shape as [241, 336, 283]
which is given as the [dept, height, width]
, then using torch.stack
should work:
a = torch.randn(241, 336, 283)
b = torch.randn(241, 336, 283)
c = torch.randn(241, 336, 283)
features = torch.stack((a, b, c), dim=0)
print(features.shape)
# torch.Size([3, 241, 336, 283])
# add batch dimension
features = features.unsqueeze(0)
print(features.shape)
# torch.Size([1, 3, 241, 336, 283])
I don’t know, but maybe @MONAI would know good resources.