Medical-SAM2

Medical-SAM2 copied to clipboard

I appreciate the excellent work of the authors, if I understand correctly./checkpoints/sam2_hiera_small.pt is a direct use of SAM2 weights, do you consider posting your own fine-tuning sam2_hiera_small.pt weights in medical...

yyy1998-i

input size

1

Thank you for your excellent work. In the paper, it is stated that: "To standardize image sizes, the provided samples were first cropped to their non - zero regions and...

yyy1998-i

hi， thanks a lot for your great work， so where is the preprocessing code for btcv dataset？

wefwefWEF2

weighted pick up strategy and 'confidence-first'

1

In learning about this excellent work, we are concerned that the memory strategy proposed in the paper has not been implemented when dealing with 3D images. May I ask the...

DF-W

Differences between MedSAM2 and SAM2 in Processing 3D Medicine

1

Hello, when presenting the comparison of MedSAM2 with various methods in the paper, I noticed that MedSAM2's segmentation metrics on the BTCV dataset are better than SAM2's. However, I saw...

Vanityplunde

Hello, how can I get a dynamic picture of 3D abdominal segmentation visualization, such as in the project?

Vanityplunde

As a new learner, I wonder about the pos weight you use in your bce loss. Does "*2" have a special influence? Thank you very much. pos_weight = torch.ones([1]).cuda(device=device)*2 criterions...

Yaostars

nothingws

net.train_propagate_in_video in 3d training

2

Hello,may I ask a question about the net.train_propagate_in_video in 3d training? I wonder how can I find the code for this method? I can find propagate_in_vedio in samvideopredictor, but I...

DemerzelWang

Dear authors, Thank you for the great open-source work. Can you possibly provide the data loader for other datasets used (such as ISIC) to better reproduce your results?

moeinheidari7829