search

NKI-AI / direct

Deep learning framework for MRI reconstruction
https://docs.aiforoncology.nl/direct
Apache License 2.0
228 stars 40 forks source link

is it possible to use your tool for denoising and resampling dicom data? #214

Closed jakubMitura14 closed 2 years ago

jakubMitura14 commented 2 years ago

Hello thanks for publishing fantastic work, s it possible to use this algorithm for dicom data (I do not have access to raw format)
is it possible to resample data to different size using your algorithm - I want to resize the images from multiple modalities in prostate MRI to single array size in order to interpolate them but I would like to avoid the interpolation artifacts related to b spline method, so in principle If I understand correctly one could sample function approximating denoised signal at different resolution than original - am I correct?

Thank you for response !

georgeyiasemis commented 2 years ago

is it possible to use this algorithm for dicom data (I do not have access to raw format) is it possible to resample data to different size using your algorithm

Hi @jakubMitura14. Unfortunately DIRECT does not support DICOM data, but it allows for flexible implementation of your own PyTorch Datasets. Please check documentation!

georgeyiasemis commented 2 years ago

I want to resize the images from multiple modalities in prostate MRI to single array size in order to interpolate them but I would like to avoid the interpolation artifacts related to b spline method, so in principle If I understand correctly one could sample function approximating denoised signal at different resolution than original - am I correct?

My apologies, but I am not sure we can help you with this question or how it's related to DIRECT.

jakubMitura14 commented 2 years ago

Thank you for your response !