mrajchl

def resample_img(itk_image, out_spacing=[2.0, 2.0, 2.0], is_label=False):
    
    # Resample images to 2mm spacing with SimpleITK
    original_spacing = itk_image.GetSpacing()
    original_size = itk_image.GetSize()

    out_size = [
        int(np.round(original_size[0] * (original_spacing[0] / out_spacing[0]))),
        int(np.round(original_size[1] * (original_spacing[1] / out_spacing[1]))),
        int(np.round(original_size[2] * (original_spacing[2] / out_spacing[2])))]

mrajchl

# Generator function
def f():
    fn = read_fn(file_references=all_filenames,
                 mode=tf.estimator.ModeKeys.TRAIN, 
                 params=reader_params)
    
    ex = next(fn)
    # Yield the next image
    yield ex
    

mrajchl

def read_fn(file_references, mode, params=None):
    
    # We define a `read_fn` and iterate through the `file_references`, which
    # can contain information about the data to be read (e.g. a file path):
    for meta_data in file_references:
        
        # Here, we parse the `subject_id` to construct a file path to read
        # an image from.
        subject_id = meta_data[0]
        data_path = '../../data/IXI_HH/1mm'

mrajchl

def decode(serialized_example):
    # Decode examples stored in TFRecord
    # NOTE: make sure to specify the correct dimensions for the images
    features = tf.parse_single_example(
        serialized_example,
        features={'train/image': tf.FixedLenFeature([128, 224, 224, 1], tf.float32),
                  'train/label': tf.FixedLenFeature([], tf.int64)})

    # NOTE: No need to cast these features, as they are already `tf.float32` values.
    return features['train/image'], features['train/label']

mrajchl

def _int64_feature(value):
    return tf.train.Feature(int64_list=tf.train.Int64List(value=[value]))

def _float_feature(value):
    return tf.train.Feature(float_list=tf.train.FloatList(value=value))

# path to save the TFRecords file
train_filename = 'train.tfrecords'

# open the file

mrajchl

import SimpleITK as sitk
import numpy as np

# A path to a T1-weighted brain .nii image:
t1_fn = './brain_t1_0001.nii'

# Read the .nii image containing the volume with SimpleITK:
sitk_t1 = sitk.ReadImage(t1_fn)

# and access the numpy array:

mrajchl

# Load all data into memory
data = load_data(all_filenames, tf.estimator.ModeKeys.TRAIN, reader_params)

# Create placeholder variables and define their shapes (here, 
# we input a volume image of size [128, 224, 244] and a single
# channel (i.e. greyscale):
x = tf.placeholder(reader_example_dtypes['features']['x'], 
                   [None, 128, 224, 224, 1])
y = tf.placeholder(reader_example_dtypes['labels']['y'], 
                   [None, 1])

mrajchl

def foo():
  print('bar')