import sys
import os
from glob2 import glob
import nibabel as nib
import numpy as np
import matplotlib
matplotlib.use('agg')
import matplotlib.pyplot as plt
from skimage.morphology import ball, binary_dilation, binary_closing
def get_foldernames(base_path, no_small=False):
walker = [x[0] for x in os.walk(base_path)]
walker.remove(base_path)
# Get subdirectory names
subdirs = [os.path.split(i) for i in walker]
# Select folders with all-numeric names
paths = [name for (i, name) in enumerate(subdirs)
if all(char.isdigit() for char in name[1])]
# Remove folders without 2 weeks of data
if no_small:
paths = [name for name in paths if len(name[1]) > 3]
return [os.path.join(p[0], p[1]) for p in paths]
def get_filenames(path, regexp):
image_path = os.path.join(path, regexp)
# Get file names
names = glob(image_path)
return names
def gen_savename(files, save_base_path, save_type, ext='.nii'):
sfiles = []
ids = []
for file in files:
# Get basepath and animal id
base_path, _ = os.path.split(file[0])
base_path, id = os.path.split(base_path)
# save_base_path = os.path.join(base_path, 'Radiomics_16bit')
sfiles.append(os.path.join(save_base_path, id + save_type + ext + '.gz'))
ids.append(id)
return sfiles, ids
def rewrite_nifti(files, sfiles):
for i in range(len(files)):
# Load file
X = nib.load(files[i][0]).get_data().astype(np.float32)
# Normalize if image
if len(np.unique(X)) > 3:
X = (X - X.min()) / (X.max() - X.min())
X = (2**16 - 1) * X
X = X.astype(np.uint16)
# Rewrite file
nib.save(nib.Nifti1Image(X, np.eye(4)), sfiles[i])
def dilate_masks(mask_files, im_files, dilate_kernel_size, diff=False):
print('Dilating tumor masks')
sfiles = []
for i in range(len(mask_files)):
print('\t%s' % mask_files[i])
# Generate updated filename
path, ext1 = os.path.splitext(mask_files[i])
path, ext2 = os.path.splitext(path)
path, filename = os.path.split(path)
sfile = os.path.join(path, filename + '_dilated' + ext2 + ext1)
# Load file
X_mask = nib.load(mask_files[i]).get_data().astype(np.bool)
X = nib.load(im_files[i][0]).get_data().astype(np.float32)
# Dilate the mask
selem = ball(dilate_kernel_size)
X_dia = binary_dilation(X_mask, selem)
# Remove air in the mask
# fig, ax = plt.subplots(1, 3, sharey=True)
# ax[0].imshow(X[:, :, 35], cmap='gray')
# ax[0].imshow(np.ma.masked_where(X_mask == 0, X_mask)[:, :, 35], alpha=0.5, cmap='summer')
# ax[0].set_title('Original mask')
#
# ax[1].imshow(X[:, :, 35], cmap='gray')
# ax[1].imshow(np.ma.masked_where(X_dia == 0, X_dia)[:, :, 35], alpha=0.5, cmap='summer')
# ax[1].set_title('Dilated mask')
#
# X_dia[X < 700] = 0
# X_dia = binary_closing(X_dia, selem=selem)
#
# ax[2].imshow(X[:, :, 35], cmap='gray')
# ax[2].imshow(np.ma.masked_where(X_dia == 0, X_dia)[:, :, 35], alpha=0.5, cmap='summer')
# ax[2].set_title('Thresholded, closed mask')
#
# plt.savefig(os.path.join(path, filename + '.png'))
# plt.close(fig)
if diff:
X_dia = np.logical_xor(X_dia, X_mask)
# Save the file
nib.save(nib.Nifti1Image(X_dia.astype(np.uint16), np.eye(4)), sfile)
sfiles.append(sfile)
return sfiles
def write_csv(csv_file, id, image_file, mask_file):
f = open(csv_file, 'w')
f.write('ID,Image,Mask\n')
for i in range(len(id)):
write_str = id[i] + ',' + image_file[i] + ',' + mask_file[i] + '\n'
f.write(write_str)
def gen_images_csv(base_path, save_base_path, dilate, ncontrasts=1, regen=True, diff_mask=False):
"""
Compiles images to be processed into a csv for radiomic processing
TODO: add multi-contrast processing
Args:
base_path (str): path to the image base directories
save_base_path (str): path in which to save images
dilate (int): mask dilation radius, no dilation is performed if dilate=0
ncontrasts (int): the number of image contrasts to use, only 1 or 3 are allowed
regen (bool): whether or not to re-write image files and masks as uint16
Returns:
csv_file (str): path to the csv file containing images to be processed
"""
# Make sure save path exists
if not os.path.exists(save_base_path):
os.mkdir(save_base_path)
# Dilation setting for masks - UPDATE VALUES
if dilate == 0:
dilate = False
dilate_rad = 0
else:
dilate = True
dilate_rad = 10
# Get subdirs of the base paths
subdirs = get_foldernames(base_path, no_small=True)
if ncontrasts == 1:
# Get filenames
image1_regexp = '*1_T2*'
image2_regexp = '*2_T2*'
label1_regexp = '*1-label.nii'
label2_regexp = '*2-label.nii'
image1 = [get_filenames(sub, image1_regexp) for sub in subdirs]
image2 = [get_filenames(sub, image2_regexp) for sub in subdirs]
label1 = [get_filenames(sub, label1_regexp) for sub in subdirs]
label2 = [get_filenames(sub, label2_regexp) for sub in subdirs]
# Get save names and ids
simage1, id = gen_savename(image1, save_base_path, save_type='_T2_1')
simage2, _ = gen_savename(image2, save_base_path, save_type='_T2_2')
slabel1, _ = gen_savename(label1, save_base_path, save_type='_mask_1')
slabel2, _ = gen_savename(label2, save_base_path, save_type='_mask_2')
# Rewrite to 16 bit
if regen:
print('Re-writing images as 16-bit')
rewrite_nifti(image1, simage1)
rewrite_nifti(image2, simage2)
rewrite_nifti(label1, slabel1)
rewrite_nifti(label2, slabel2)
# Dilate masks
if dilate:
slabel1 = dilate_masks(slabel1, image1, dilate_kernel_size=dilate_rad, diff=diff_mask)
slabel2 = dilate_masks(slabel2, image2, dilate_kernel_size=dilate_rad, diff=diff_mask)
# Write CSV file
csv_file = os.path.join(save_base_path, 'radiomics_files.csv')
write_csv(csv_file, 2 * id, simage1 + simage2, slabel1 + slabel2)
elif ncontrasts == 3:
# Get filenames
image1T1_regexp = '*1_T1.nii'
image1T1c_regexp = '*1_T1*C*'
image1T2_regexp = '*1_T2*'
image2T1_regexp = '*2_T1.nii'
image2T1c_regexp = '*2_T1*C*'
image2T2_regexp = '*2_T2*'
label1_regexp = '*1-label.nii'
label2_regexp = '*2-label.nii'
image1T1 = [get_filenames(sub, image1T1_regexp) for sub in subdirs]
image1T1c = [get_filenames(sub, image1T1c_regexp) for sub in subdirs]
image1T2 = [get_filenames(sub, image1T2_regexp) for sub in subdirs]
image2T1 = [get_filenames(sub, image2T1_regexp) for sub in subdirs]
image2T1c = [get_filenames(sub, image2T1c_regexp) for sub in subdirs]
image2T2 = [get_filenames(sub, image2T2_regexp) for sub in subdirs]
label1 = [get_filenames(sub, label1_regexp) for sub in subdirs]
label2 = [get_filenames(sub, label2_regexp) for sub in subdirs]
# Get save names and ids
simage1T1, id = gen_savename(image1T1, save_base_path, save_type='_T1_1')
simage1T1c, _ = gen_savename(image1T1c, save_base_path, save_type='_T1c_1')
simage1T2, _ = gen_savename(image1T2, save_base_path, save_type='_T2_1')
simage2T1, _ = gen_savename(image2T1, save_base_path, save_type='_T1_2')
simage2T1c, _ = gen_savename(image2T1c, save_base_path, save_type='_T1c_2')
simage2T2, _ = gen_savename(image2T2, save_base_path, save_type='_T2_2')
slabel1, _ = gen_savename(label1, save_base_path, save_type='_mask_1')
slabel2, _ = gen_savename(label2, save_base_path, save_type='_mask_2')
# Rewrite to 16 bit
if regen:
print('Re-writing images as 16-bit')
rewrite_nifti(image1T1, simage1T1)
rewrite_nifti(image1T1c, simage1T1c)
rewrite_nifti(image1T2, simage1T2)
rewrite_nifti(image2T1, simage2T1)
rewrite_nifti(image2T1c, simage2T1c)
rewrite_nifti(image2T2, simage2T2)
rewrite_nifti(label1, slabel1)
rewrite_nifti(label2, slabel2)
# Dilate masks
if dilate and regen:
slabel1 = dilate_masks(slabel1, image1T2, dilate_kernel_size=dilate_rad, diff=diff_mask)
slabel2 = dilate_masks(slabel2, image2T2, dilate_kernel_size=dilate_rad, diff=diff_mask)
# Write CSV file
csv_file = os.path.join(save_base_path, 'radiomics_files_multi.csv')
write_csv(csv_file,
6 * id, simage1T1 + simage1T1c + simage1T2 + simage2T1 + simage2T1c + simage2T2,
3*slabel1 + 3*slabel2)
else:
raise ValueError('INVALID NUMBER OF CONTRASTS: Enter 1 or 3 contrasts')
return csv_file
def gen_images_csv_t1c(base_path, save_base_path, dilate, ncontrasts=1, regen=True, diff_mask=False):
"""
Compiles images to be processed into a csv for radiomic processing
TODO: add multi-contrast processing
Args:
base_path (str): path to the image base directories
save_base_path (str): path in which to save images
dilate (int): mask dilation radius, no dilation is performed if dilate=0
ncontrasts (int): the number of image contrasts to use, only 1 or 3 are allowed
regen (bool): whether or not to re-write image files and masks as uint16
Returns:
csv_file (str): path to the csv file containing images to be processed
"""
# Make sure save path exists
if not os.path.exists(save_base_path):
os.mkdir(save_base_path)
# Dilation setting for masks - UPDATE VALUES
if dilate == 0:
dilate = False
dilate_rad = 0
else:
dilate = True
dilate_rad = 10
# Get subdirs of the base paths
subdirs = get_foldernames(base_path, no_small=True)
# Get filenames
image1_regexp = '*1_T1*C*'
image2_regexp = '*2_T1*C*'
label1_regexp = '*1-label.nii'
label2_regexp = '*2-label.nii'
image1 = [get_filenames(sub, image1_regexp) for sub in subdirs]
image2 = [get_filenames(sub, image2_regexp) for sub in subdirs]
label1 = [get_filenames(sub, label1_regexp) for sub in subdirs]
label2 = [get_filenames(sub, label2_regexp) for sub in subdirs]
# Get save names and ids
simage1, id = gen_savename(image1, save_base_path, save_type='_T1C_1')
simage2, _ = gen_savename(image2, save_base_path, save_type='_T1C_2')
slabel1, _ = gen_savename(label1, save_base_path, save_type='_mask_1')
slabel2, _ = gen_savename(label2, save_base_path, save_type='_mask_2')
# Rewrite to 16 bit
if regen:
print('Re-writing images as 16-bit')
rewrite_nifti(image1, simage1)
rewrite_nifti(image2, simage2)
rewrite_nifti(label1, slabel1)
rewrite_nifti(label2, slabel2)
# Dilate masks
if dilate:
slabel1 = dilate_masks(slabel1, image1, dilate_kernel_size=dilate_rad, diff=diff_mask)
slabel2 = dilate_masks(slabel2, image2, dilate_kernel_size=dilate_rad, diff=diff_mask)
# Write CSV file
csv_file = os.path.join(save_base_path, 'radiomics_files.csv')
write_csv(csv_file, 2 * id, simage1 + simage2, slabel1 + slabel2)
return csv_file
if __name__ == "__main__":
# Dilation setting for masks - UPDATE VALUES
dilate = True
dilate_rad = 10
# Paths to base data (float format)
if sys.platform == 'linux':
base_path = '/media/matt/Seagate Expansion Drive/MR Data/MR_Images_Sarcoma'
save_base_path = '/media/matt/Seagate Expansion Drive/MR Data/MR_Images_Sarcoma/Radiomics_16bit'
else:
base_path = 'E:\\MR Data\\MR_Images_Sarcoma'
save_base_path = 'E:\\MR Data\\MR_Images_Sarcoma\\Radiomics_16bit'
# Get subdirs of the base paths
subdirs = get_foldernames(base_path, no_small=True)
# Get filenames
image1_regexp = '*1_T2*'
image2_regexp = '*2_T2*'
label1_regexp = '*1-label.nii'
label2_regexp = '*2-label.nii'
image1 = [get_filenames(sub, image1_regexp) for sub in subdirs]
image2 = [get_filenames(sub, image2_regexp) for sub in subdirs]
label1 = [get_filenames(sub, label1_regexp) for sub in subdirs]
label2 = [get_filenames(sub, label2_regexp) for sub in subdirs]
# Get save names and ids
simage1, id = gen_savename(image1, save_base_path, save_type='_T2_1')
simage2, _ = gen_savename(image2, save_base_path, save_type='_T2_2')
slabel1, _ = gen_savename(label1, save_base_path, save_type='_mask_1')
slabel2, _ = gen_savename(label2, save_base_path, save_type='_mask_2')
# Rewrite to 16 bit
rewrite_nifti(image1, simage1)
rewrite_nifti(image2, simage2)
rewrite_nifti(label1, slabel1)
rewrite_nifti(label2, slabel2)
# Dilate masks
if dilate:
slabel1 = dilate_masks(slabel1, image1, dilate_kernel_size=dilate_rad)
slabel2 = dilate_masks(slabel2, image2, dilate_kernel_size=dilate_rad)
# Write CSV file
csv_file = os.path.join(save_base_path, 'radiomics_files.csv')
write_csv(csv_file, 2*id, simage1 + simage2, slabel1 + slabel2)
# write_csv(csv_file, id[:2], simage1[:2], slabel1[:2])
