search

MIC-DKFZ / nnUNet

Apache License 2.0
5.95k stars 1.77k forks source link

Small object segmentation #2532

Open davidkflau opened 1 month ago

davidkflau commented 1 month ago

Hi team,

I'm training on brain lesion segmentation task. The lesions have various sizes, from <10mL to >200mL. We first feed the small and large lesion data to the nnUNet, but the average dice is not good, especially the dice of small lesions are bad. We then tried to train second model with large lesion data only and the dice are much better than the first model.

Just wonder if nnUNet works well on small object, and do you have any suggestion to segment small and large objects at the same time?

Thanks.

fitzjalen commented 1 month ago

Hello In my opinion patch size might play role and maybe Encoder Could you please share your training plans?

davidkflau commented 1 month ago

plans.json

Here's the training plan. I also tried the EncoderM architecture, but it doesn't help much.

fitzjalen commented 1 month ago

plans.json

Here's the training plan. I also tried the EncoderM architecture, but it doesn't help much.

One more important question I forgot to ask: Did you count small segments in dataset ? I mean the number of separate segments with size less than 100 voxels for example

TaWald commented 2 weeks ago

Is the issue is still persisting? If not it would be great if you @davidkflau would post your solution and close the issue for others that have similar questions