You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello, thanks for the pre-print and releasing the code. I am working on brain tumor segmentation with the BraTS dataset, which is semantic, as you know. I have two questions that would be nice to have some clarity on.
What is actually being evaluated in the pre-print? The objectives seem like they're binary, such as the dataset dataset/brat.py seeming to show it doing binary segmentation (segmentation mask is binarized). I know Segment Anything is natively a binary segmentation algorithm, but how come there are class-wise results for the BTCV organ dataset and not BraTS? @LJQCN101 supposedly integrated multimask output for semantic segmentation, but the results in the pre-print don't seem to reflect this.
Does medical SAM by default use channel wise segmentation with multiple modalities (e.g. in the case of brain tumors, flair, t1, t1ce, t2) or does it repeat the same modality across multiple channels? I ask this because in the dataset it shows only the first level being used.
I am adding the authors of the pre-print and the person who implemented multimask output below. Thanks again! @WuJunde@LJQCN101
The text was updated successfully, but these errors were encountered:
Hello, thanks for the pre-print and releasing the code. I am working on brain tumor segmentation with the BraTS dataset, which is semantic, as you know. I have two questions that would be nice to have some clarity on.
What is actually being evaluated in the pre-print? The objectives seem like they're binary, such as the dataset
dataset/brat.py
seeming to show it doing binary segmentation (segmentation mask is binarized). I know Segment Anything is natively a binary segmentation algorithm, but how come there are class-wise results for the BTCV organ dataset and not BraTS? @LJQCN101 supposedly integrated multimask output for semantic segmentation, but the results in the pre-print don't seem to reflect this.Does medical SAM by default use channel wise segmentation with multiple modalities (e.g. in the case of brain tumors, flair, t1, t1ce, t2) or does it repeat the same modality across multiple channels? I ask this because in the dataset it shows only the first level being used.
I am adding the authors of the pre-print and the person who implemented multimask output below. Thanks again!
@WuJunde @LJQCN101
The text was updated successfully, but these errors were encountered: