search

Julian-Wyatt / AnoDDPM

CVPR Workshop paper - AnoDDPM: Anomaly Detection with Denoising Diffusion Probabilistic Models using Simplex Noise
https://julianwyatt.co.uk/anoddpm
MIT License
174 stars 28 forks source link

Which file to use exactly for the evaluation of MRI dataset? #4

Closed ridhimagarg closed 2 years ago

ridhimagarg commented 2 years ago

Hi!

I was looking for how to use detection.py, In that we have to put the dataset in ./DATASETS/CancerousDataset/EdinburghDataset/Anomalous-T1/raw but in raw folder for all 22 scans which file we have to keep, there are many files present in the one scan. Should it be binary file in Axial_T2 folder for each scan?

Some guidance is helpful!

Thanks!

Julian-Wyatt commented 2 years ago

Hi, my specific pipeline, converted the .nii files to .npy files as they are much quicker to read into PyTorch. The function I used to automate this is main() in dataset.py.

These resultant .npy files are those placed in that 'raw' folder. You can however structure your folders however you see fit.

Please let me know if this helps.

ridhimagarg commented 2 years ago

Thanks for your response!!

Yes, but I was thinking if which images you are utilising in your original paper for evaluating as for each scan there are multiple folders which contain .nii files. Like for this scan -: image Which folder should I use?

Julian-Wyatt commented 2 years ago

Ah yes - wonderful question because that really was confusing.

So I went through all of those .nii files and used the COR_3D_IR images, which I believe are the T1 weighted images from that dataset.

Does this help?

ridhimagarg commented 2 years ago

Thanks a lot!!