search

qinliuliuqin / Medical-Detection3d-Toolkit

Landmark detection engine for 3D medical images (MICCAI workshop 2021)
GNU General Public License v3.0
35 stars 8 forks source link

Preprocessing #1

Closed cadriaans closed 7 months ago

cadriaans commented 1 year ago

Hi, i was wondering if preprocessing of the data is included in this model or if you need to preprocess the ct data yourself? Like resampling, reshape, alignment and normalization?

sunshinelike commented 11 months ago

How to get the landmark.csv with every data in the preprocessing phrase? Looking forward to your reply. Thank you very much.

qinliuliuqin commented 7 months ago

@sunshinelike Hi, sorry for the late response. I overlooked this issue. An example landmark csv file can be found here: https://github.com/qinliuliuqin/Medical-Detection3d-Toolkit/blob/master/demo/case_17_cbct_patient.csv. You need to generate these files by yourself.

qinliuliuqin commented 7 months ago

@cadriaans Hi, during training, we will crop, normalize, and resample the image. https://github.com/qinliuliuqin/Medical-Detection3d-Toolkit/blob/7a7d98debe54604e38cfd5fad8cb50fdd8d028be/detection3d/config/lmk_train_config.py#L56. During inference, we take the whole image as input and resample it to the training spacing, and then normalize it. More details please refer to the code.