A question about processing

xmengli / H-DenseUNet

TMI 2018. H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation from CT Volumes

532 stars 161 forks source link

Open Kristin-ZLCHEN opened 5 years ago

Kristin-ZLCHEN commented 5 years ago

Should the original segmentation-.nii be processed before training?

DanielLee343 commented 5 years ago

I think segmentation and volume are pretty much like key-value format, so they shall be preprocessed in pairs.

zz10001 commented 4 years ago

Hi， I notice the paper resample when liver segmnetation, while use orginal resolution(don't resample) when liver lesion segmentation, Is it right?

xmengli commented 4 years ago

Yes

zz10001 commented 4 years ago

Thanks for your kind help!