def isdicom(path):
""" Judge whether a given file is a valid dicom.
Args:
path(str): given file path.
Returns:
(bool): True if given file path is a valid dicom, otherwise False.
"""
if not mv.isfile(path):
return False
# read preamble and magic code
with open(path, 'rb') as f:
header = f.read(132)
if not header:
return False
# magic code of a dicom file should be 'DICM'
magic_code = header[128:132]
if magic_code != b'DICM':
return False
else:
return True
def compute_md5_str(file_path):
if not mv.isfile(file_path):
return None
with open(file_path, 'rb') as f:
m = hashlib.md5()
m.update(f.read())
md5_code = m.hexdigest()
return str(md5_code).lower()
def __init__(self, cfg, mode, build_transform, image_loader):
self.is_train = (mode == mv.ModeKey.TRAIN)
self.mode2dsmd = {
mv.ModeKey.TRAIN: cfg.DATA.TRAIN_DSMD,
mv.ModeKey.VAL: cfg.DATA.VAL_DSMD,
mv.ModeKey.TEST: cfg.DATA.TEST_DSMD,
}
dsmd_path = self.mode2dsmd[mode]
assert mv.isfile(dsmd_path) or mv.isdir(dsmd_path)
if mv.isfile(dsmd_path): # for a dsmd file
self.dsmd = mv.load_dsmd(dsmd_path)
self.filenames = list(self.dsmd.keys())
self.filepaths = [
mv.joinpath(cfg.DATA.IMAGE_DIR, filename)
for filename in self.filenames
]
else: # for a directory containing test images
self.dsmd = None
self.filepaths = mv.listdir(dsmd_path)
self.transform = build_transform(cfg, self.is_train)
self.image_loader = image_loader
def test_split_dsmd_file(dsmd_file):
tmp_dir = mv.joinpath(DATA_DIR, 'temporary_subdir')
tmp_path = mv.joinpath(tmp_dir, 'tmp_dsmd.txt')
mv.mkdirs(tmp_dir)
mv.cp(dsmd_file, tmp_path)
datasplit = {'train': 0.9, 'val': 0.1, 'test': 0.0}
# shuffle
mv.split_dsmd_file(tmp_path, datasplit)
train_dsmd_file_path = mv.joinpath(tmp_dir, 'train.csv')
val_dsmd_file_path = mv.joinpath(tmp_dir, 'val.csv')
test_dsmd_file_path = mv.joinpath(tmp_dir, 'test.csv')
assert mv.isfile(train_dsmd_file_path)
assert mv.isfile(val_dsmd_file_path)
assert not mv.isfile(test_dsmd_file_path)
train_dsmd = mv.load_dsmd(train_dsmd_file_path)
val_dsmd = mv.load_dsmd(val_dsmd_file_path)
assert len(train_dsmd) == 18
assert len(val_dsmd) == 2
# non shuffle
mv.split_dsmd_file(tmp_path, datasplit, shuffle=False)
train_dsmd_file_path = mv.joinpath(tmp_dir, 'train.csv')
val_dsmd_file_path = mv.joinpath(tmp_dir, 'val.csv')
test_dsmd_file_path = mv.joinpath(tmp_dir, 'test.csv')
assert mv.isfile(train_dsmd_file_path)
assert mv.isfile(val_dsmd_file_path)
assert not mv.isfile(test_dsmd_file_path)
train_dsmd = mv.load_dsmd(train_dsmd_file_path)
val_dsmd = mv.load_dsmd(val_dsmd_file_path)
assert len(train_dsmd) == 18
assert len(val_dsmd) == 2
assert 'brain_001.dcm' in train_dsmd
assert 'brain_019.dcm' in val_dsmd
mv.rmtree(tmp_dir)