Modality-Agnostic Learning for Medical Image Segmentation Using Multi-modality Self-distillation
APACHE-2.0 License
Bot releases are visible (Hide)
API updates:
None
nn.FeaturedData
and nn.MAGNET2
builder
packageMid*Fusion
module to Mean*Fusion
losses.MAGLoss
when a modality is not givenOther updates:
Published by kisonho 9 months ago
Updates:
Published by kisonho 9 months ago
API updates:
None
nn.FeaturedData
and nn.MAGNET2
builder
packageMid*Fusion
module to Mean*Fusion
losses.MAGLoss
when a modality is not givenBeta updates:
Published by kisonho 10 months ago
API updates:
None
nn.FeaturedData
and nn.MAGNET2
builder
packageMid*Fusion
module to Mean*Fusion
losses.MAGLoss
when a modality is not givenBeta updates:
Published by kisonho 10 months ago
API updates:
None
nn.FeaturedData
and nn.MAGNET2
builder
packageMid*Fusion
module to Mean*Fusion
losses.MAGLoss
when a modality is not givenOther updates:
Published by kisonho 11 months ago
API Updates:
torchmanager_monai.Manager
as general manager for monailosses.MAGMSLoss
MonaiManager
torchmanager_monai
as its extra package (MonaiManager
, networks.UNETR
, and networks.UNETRWithMultiModality
will be defined as NotImplemented
)magnet.managers.monai.Manager
to magnet.manager.monai.SegmentationManager
torchmanager_monai.MonaiManager
to SegmentationManager
which inherits torchmanager_monai.Monai
Other updates:
losses.protocols.DistillatedData
Published by kisonho 12 months ago
Updates:
return_features
in build_v2_unet
function to True
Published by kisonho about 1 year ago
API Updates:
torchmanager_monai.Manager
as general manager for monailosses.MAGMSLoss
torchmanager_monai
as its extra package (MonaiManager
, networks.UNETR
, and networks.UNETRWithMultiModality
will be defined as NotImplemented
)magnet.managers.monai.Manager
to magnet.manager.monai.SegmentationManager
torchmanager_monai.MonaiManager
to SegmentationManager
which inherits torchmanager_monai.Monai
Other updates:
losses.protocols.DistillatedData
Published by kisonho about 1 year ago
Updates:
losses.protocols.DistillatedData
Published by kisonho about 1 year ago
API Updates:
torchmanager_monai.Manager
as general manager for monailosses.MAGMSLoss
torchmanager_monai
as its extra package (MonaiManager
, networks.UNETR
, and networks.UNETRWithMultiModality
will be defined as NotImplemented
)magnet.managers.monai.Manager
to magnet.manager.monai.SegmentationManager
torchmanager_monai.MonaiManager
to SegmentationManager
which inherits torchmanager_monai.Monai
Other updates:
MonaiManager
Published by kisonho about 1 year ago
Feature updates:
losses.MAGFeatureDistillationLoss
and losses.MAGSelfDistillationLoss
)copy_encoder
boolean flag to control if use the same initialized weights for modality specific encodersMAGNET2
architecture with feature fusionHeMIS
when target_dict
does not give all available modalities but inputs contain all the modalitiesmagms
target
property can now be set as a list
of target modalities in both TargetingManager
and MAGNET
Published by kisonho about 1 year ago
Updates:
Published by kisonho about 1 year ago
list
of torch.Tensor
when aggregating a CumulativeIteration
monai metric.Published by kisonho over 1 year ago
Updates:
Published by kisonho over 1 year ago
API updates:
torch.data.Dataset
and torch.data.DataLoader
can now be used to train with all modalities inputs instead of using magnet.data.TargetingDataset
and magnet.data.TargetingDataLoader
copy_modality
boolean flag control for build
methodHeMIS
network implementation as a comparisonUNETRWithDictOutput
for dictionary outputsmagnet.losses
packages where magnet.losses.MAGLoss
can be used to combine the training of Magnet
in one iterationnetworks.unetr
packageTargeting
protocol now accepts optional keys in target_dict
property for all modalitiestarget
property can now be set as a list
of target modalities in both TargetingManager
and MAGNET
Published by kisonho over 1 year ago
Updates:
Published by kisonho over 1 year ago
Updates:
Published by kisonho almost 2 years ago
Updates:
TargetingManager
Published by kisonho almost 2 years ago
Main release for MAGNET for ISBI 2023 review