Hello everyone,
I have a couple of questions regarding the procedure for AI segmentation using UNet for ultrasound segmentation.
Firstly, is there any way to export the segmentation from the single slide segmentation directly to an STL file? The reason I find this preferable to the normal segmentation module is due to the speed of segmentation in the single slide module. It also allows me to easily perform segmentation on 2D images and generate a 3D model.
Secondly, I am following the AI training protocol in the README file of the ultrasound segmentation on the AIGT GitHub. However, it doesn’t provide me with the .h5 file needed for my next automatic volume segmentation. How can I obtain this file?
Lastly, when I perform segmentation on a specimen for AI training and export the data to work on the AI procedure, how should I update this trained data using new samples? For example, if I have 5 different samples, how should I incorporate them?
I appreciate any guidance on these questions.
Thank you.