search

ginobilinie / medSynthesis

medical image synthesis with GAN framework based on tensorflow (we also have a better version in pytorch: https://github.com/ginobilinie/medSynthesisV1)
MIT License
72 stars 24 forks source link

The generated result is not desirable,what's the possible reason? #11

Closed huan-zi closed 3 years ago

huan-zi commented 5 years ago

I 'm trying to generate CT from MRI with 3D vision, but the result doesn't look like CT. It's like a label map,no details.And the loss is already changing a little.Have you met this problem?

ginobilinie commented 5 years ago

I suggest you use big patch, something like 144x144x16. It is relatively harder to train the model using small 3d patch.

ginobilinie commented 5 years ago

Also, I strongly suggest you use another copy of my code: https://github.com/ginobilinie/medSynthesisV1