Application to semantic segmentation

[drusmanbashir]

Hi,
Thank you for providing such a powerful tool for data-efficient training! I’ve been working on a medical imaging project focused on liver lesion segmentation in patients with colorectal cancer. Our dataset includes 220 CT scans, and we are trying to optimize the training process by discarding non-contributory data points using a sampling strategy like CRAIG.

I’ve explored DeepCore for classification tasks but am unsure how to best adapt it for semantic segmentation, where pixel-level accuracy is crucial. Could you please advise on how the code might be modified to handle segmentation tasks or provide guidance on integrating DeepCore for such use cases?

I’d greatly appreciate any help or suggestions!