Closed skill51988 closed 9 months ago
Hi @skill51988 ,
Thanks for your interest very much.
The low performance in the z-axis is expected since the training data didn't contain data in z-axis.
Nevertheless, the model can be easily fine-tuned to the new data. here is an example. https://github.com/bowang-lab/MedSAM/blob/LiteMedSAM/README.md
Hi,
I am currently working on characterising the human eye using T1 MRI images. It seems to successfully segment the eye in the axial plane, yet the outcome in the z-axis is not ideal
I wonder whether this plug-in is trainable and if you have any suggestions.
Thanks!!