Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Register data #4

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Register data #4

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
71 changes: 49 additions & 22 deletions register_data.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,42 +22,69 @@ trap "echo Caught Keyboard Interrupt within script. Exiting now.; exit" INT

# Retrieve input params
SUBJECT=$1
FILEPARAM=$2
#FILEPARAM=$2

# SCRIPT STARTS HERE
# ==============================================================================
# shellcheck disable=SC1090
source $FILEPARAM
#source $FILEPARAM
# Go to results folder, where most of the outputs will be located
cd $PATH_RESULTS
cd $PATH_RESULTS;
# Copy source images and segmentations
mkdir -p data/derivatives/labels
cd data
mkdir -p data/derivatives/labels;
cd data;

cp -r $PATH_DATA/$SUBJECT .
cp -r $PATH_DATA/derivatives/labels/$SUBJECT $PATH_RESULTS/data/derivatives/labels
cp -r $PATH_DATA/$SUBJECT .;
cp -r $PATH_DATA/derivatives/labels/$SUBJECT $PATH_RESULTS/data/derivatives/labels;

# Go to data folder
cd $PATH_RESULTS/data/$SUBJECT/anat/
cd $PATH_RESULTS/data/$SUBJECT/anat/;
## Setup file names
file_t2w=${SUBJECT}_T2w
file_t1w=${SUBJECT}_T1w
path_t1w=`pwd`/${file_t1w}.nii.gz
file_t2w=${SUBJECT}_T2w;
file_t1w=${SUBJECT}_T1w;
path_t1w=`pwd`/${file_t1w}.nii.gz;

if [ -f ${path_t1w} ]; then
sct_register_multimodal -i ${file_t1w}.nii.gz -d ${file_t2w}.nii.gz -x spline
mv ${file_t2w}_reg.nii.gz ${file_t2w}.nii.gz
mv ${file_t1w}_reg.nii.gz ${file_t1w}.nii.gz
warp_T2w2T1w=`pwd`/warp_${file_t2w}2${file_t1w}.nii.gz
warp_T1w2T2w=`pwd`/warp_${file_t1w}2${file_t2w}.nii.gz
# sct_resample -i ${file_t1w}.nii.gz -mm 0.2x0.2x0.5 -o ${file_t1w}.nii.gz;
# sct_resample -i ${file_t2w}.nii.gz -mm 0.2x0.2x0.5 -o ${file_t2w}.nii.gz;
# sct_crop_image -i ${file_t1w}.nii.gz -ref ${file_t2w}.nii.gz -o ${file_t1w}.nii.gz;
# sct_resample -i ${file_t1w}.nii.gz -mm 0.5x0.5x2 -o ${file_t1w}.nii.gz;
# sct_resample -i ${file_t2w}.nii.gz -mm 0.5x0.5x2 -o ${file_t2w}.nii.gz;
# sct_resample -i ${path_mask}${file_t2w}_centerline.nii.gz -mm 0.5x0.5x2 -o ${path_mask}${file_t2w}_centerline.nii.gz;
# step below to improve with dedicated optic model
sct_get_centerline -i ${file_t1w}.nii.gz -c t1 -method optic;
sct_create_mask -i ${file_t1w}.nii.gz -size 30mm -f cylinder -p centerline,${file_t1w}_centerline.nii.gz -o mask.nii.gz;
# sct_crop_image -i ${file_t1w}.nii.gz -m mask.nii.gz -o ${file_t1w}.nii.gz;
# sct_crop_image -i ${file_t2w}.nii.gz -m mask.nii.gz -o ${file_t2w}.nii.gz;
sct_register_multimodal -i ${file_t1w}.nii.gz -d ${file_t2w}.nii.gz -m mask.nii.gz -param step=1,type=im,algo=slicereg,metric=CC;
#
mv ${file_t2w}_reg.nii.gz ${file_t2w}.nii.gz;
mv ${file_t1w}_reg.nii.gz ${file_t1w}.nii.gz;
warp_T2w2T1w=`pwd`/warp_${file_t2w}2${file_t1w}.nii.gz;
warp_T1w2T2w=`pwd`/warp_${file_t1w}2${file_t2w}.nii.gz;
fi

cd $PATH_RESULTS/data/derivatives/labels/$SUBJECT/anat/
cd $PATH_RESULTS/data/derivatives/labels/$SUBJECT/anat/;

file_t2w_seg=${SUBJECT}_T2w_seg-tumor
file_t1w_seg=${SUBJECT}_T1w_seg-tumor
file_t2w_seg=${SUBJECT}_T2w_seg-tumor;
file_t1w_seg=${SUBJECT}_T1w_seg-tumor;
file_t2w_edema=${SUBJECT}_T2w_edema;
file_t1w_edema=${SUBJECT}_T1w_edema;
path_edema=`pwd`/${file_t2w_edema}.nii.gz;

if [ -f ${path_t1w} ]; then
sct_apply_transfo -i ${file_t1w_seg}.nii.gz -d ${file_t2w_seg}.nii.gz -w ${warp_T1w2T2w} -o ${file_t1w_seg}.nii.gz -x linear
sct_apply_transfo -i ${file_t2w_seg}.nii.gz -d ${file_t1w_seg}.nii.gz -w ${warp_T2w2T1w} -o ${file_t2w_seg}.nii.gz -x linear
fi
# sct_resample -i ${file_t1w_seg}.nii.gz -mm 0.2x0.2x0.5 -o ${file_t1w_seg}.nii.gz;
# sct_crop_image -i ${file_t1w_seg}.nii.gz -ref $PATH_RESULTS/data/$SUBJECT/anat/${file_t2w}.nii.gz -o ${file_t1w_seg}.nii.gz;
# sct_resample -i ${file_t2w_seg}.nii.gz -mm 0.2x0.2x0.5 -o ${file_t2w_seg}.nii.gz;
# sct_crop_image -i ${file_t2w_seg}.nii.gz -ref $PATH_RESULTS/data/$SUBJECT/anat/${file_t2w}.nii.gz -o ${file_t2w_seg}.nii.gz;
# sct_resample -i ${file_t2w_seg}.nii.gz -mm 0.5x0.5x2 -o ${file_t2w_seg}.nii.gz;
# sct_resample -i ${file_t1w_seg}.nii.gz -mm 0.5x0.5x2 -o ${file_t1w_seg}.nii.gz;
sct_apply_transfo -i ${file_t1w_seg}.nii.gz -d ${file_t2w_seg}.nii.gz -w ${warp_T1w2T2w} -o ${file_t1w_seg}.nii.gz -x linear;
sct_apply_transfo -i ${file_t2w_seg}.nii.gz -d ${file_t1w_seg}.nii.gz -w ${warp_T2w2T1w} -o ${file_t2w_seg}.nii.gz -x linear;
if [ -f ${path_edema} ]; then
sct_apply_transfo -i ${file_t1w_edema}.nii.gz -d ${file_t2w_edema}.nii.gz -w ${warp_T1w2T2w} -o ${file_t1w_edema}.nii.gz -x linear;
sct_apply_transfo -i ${file_t2w_edema}.nii.gz -d ${file_t1w_edema}.nii.gz -w ${warp_T2w2T1w} -o ${file_t2w_edema}.nii.gz -x linear;
# sct_resample -i ${file_t2w_edema}.nii.gz -mm 0.5x0.5x2 -o ${file_t2w_edema}.nii.gz;
# sct_resample -i ${file_t1w_edema}.nii.gz -mm 0.5x0.5x2 -o ${file_t1w_edema}.nii.gz;
fi
fi
83 changes: 83 additions & 0 deletions verifyDim.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,83 @@
from __future__ import print_function

import nibabel as nib
import numpy as np
from torch import nn
from bids_neuropoly import bids
from spinalcordtoolbox.image import Image
from shutil import copyfile
import os
from ivadomed.transforms import CenterCrop

class DiceLoss(nn.Module):
"""DiceLoss.

.. seealso::
Milletari, Fausto, Nassir Navab, and Seyed-Ahmad Ahmadi. "V-net: Fully convolutional neural networks for
volumetric medical image segmentation." 2016 fourth international conference on 3D vision (3DV). IEEE, 2016.

Args:
smooth (float): Value to avoid division by zero when images and predictions are empty.

Attributes:
smooth (float): Value to avoid division by zero when images and predictions are empty.
"""
def __init__(self, smooth=1.0):
super(DiceLoss, self).__init__()
self.smooth = smooth

def forward(self, prediction, target):
iflat = prediction.reshape(-1)
tflat = target.reshape(-1)
intersection = (iflat * tflat).sum()

return - (2.0 * intersection + self.smooth) / (iflat.sum() + tflat.sum() + self.smooth)

# rootDataPath = "/home/andreanne/Documents/dataset/tumor_segmentation_masks/results/data"
rootDataPath = "/home/andreanne/Documents/dataset/tumor_segmentation_new_edema/results/data"
bids_ds = bids.BIDS(rootDataPath)
df = bids.BIDS(rootDataPath).participants.content
subject_lst = df['participant_id'].tolist()
contrast_lst = ["T2w"]
bids_subjects = [s for s in bids_ds.get_subjects() if s.record["subject_id"] in subject_lst]
max_dim = 0
filename = 0
dice_sum = 0
counter = 0
dirty = 0
h_ = 0
w_ = 0
d_ = 0

for s in subject_lst:
img_shape = None
for subject in bids_subjects:
files = []
if subject.record["subject_id"] == s:
if subject.record["modality"] in ["T2w"]:
input_file = subject.record.absolute_path
files.append(input_file)
files.append(input_file.replace("T2w", "T1w"))
derivatives = subject.get_derivatives("labels")
for deriv in derivatives:
if deriv.endswith(subject.record["modality"] + "_edema.nii.gz") or deriv.endswith(subject.record["modality"] + "_seg-tumor.nii.gz"):
files.append(deriv)
files.append(deriv.replace("T2w", "T1w"))
data = []
for file in files:
if os.path.exists(file):
input_nii = nib.load(file)
data.append(input_nii.get_fdata().shape)

if len(set(data)) != 1:
# print(files[0])
transform = CenterCrop(size=data[0])
for i in files:
if os.path.exists(i):
img_nii = nib.load(i)
img = img_nii.get_fdata()
cropped, _ = transform(img, {'crop_params': {}})
print(cropped.shape)
nib_image = nib.Nifti1Image(cropped, img_nii.affine)
nib.save(nib_image, i)