def test_masked_transformation():
image1 = nibabel.load(get_pair_images()[0])
brain_4mm = get_standard_mask(4)
nonzero_voxels = brain_4mm.get_data()[brain_4mm.get_data()!=0].shape[0]
image1_vector = make_resampled_transformation_vector(image1,resample_dim=[4,4,4],standard_mask=True)
assert_equal(nonzero_voxels,len(image1_vector))
brain_8mm = get_standard_mask(8)
nonzero_voxels = brain_8mm.get_data()[brain_8mm.get_data()!=0].shape[0]
image1_vector = make_resampled_transformation_vector(image1,resample_dim=[8,8,8],standard_mask=True)
assert_equal(nonzero_voxels,len(image1_vector))
def test_masked_transformation():
image1 = nibabel.load(get_pair_images()[0])
brain_4mm = get_standard_mask(4)
nonzero_voxels = brain_4mm.get_data()[brain_4mm.get_data() != 0].shape[0]
image1_vector = make_resampled_transformation_vector(
image1, resample_dim=[4, 4, 4], standard_mask=True)
assert_equal(nonzero_voxels, len(image1_vector))
brain_8mm = get_standard_mask(8)
nonzero_voxels = brain_8mm.get_data()[brain_8mm.get_data() != 0].shape[0]
image1_vector = make_resampled_transformation_vector(
image1, resample_dim=[8, 8, 8], standard_mask=True)
assert_equal(nonzero_voxels, len(image1_vector))
def make_resampled_transformation(nii_obj,resample_dim=[4,4,4],standard_mask=True):
nii_obj = get_nii_obj(nii_obj)[0]
# To set 0s to nan, we need to have float64 data type
true_zeros = numpy.zeros(nii_obj.shape) # default data_type is float64
true_zeros[:] = nii_obj.get_data()
true_zeros[true_zeros==0] = numpy.nan
# Resample image to 4mm voxel, nans are preserved
true_zeros = nib.nifti1.Nifti1Image(true_zeros,affine=nii_obj.get_affine())
# Standard brain masking
if standard_mask == True:
standard = get_standard_mask(voxdim=resample_dim[0])
true_zeros = resample_img(true_zeros,target_affine=standard.get_affine(),
target_shape=standard.shape)
# Mask the image
masked_true_zeros = numpy.zeros(true_zeros.shape)
masked_true_zeros[standard.get_data()!=0] = true_zeros.get_data()[standard.get_data()!=0]
true_zeros = nib.nifti1.Nifti1Image(masked_true_zeros,affine=true_zeros.get_affine())
# or just resample
else:
if (resample_dim != numpy.diag(true_zeros.get_affine())[0:3]).all():
true_zeros = resample_img(true_zeros,target_affine=numpy.diag(resample_dim))
return true_zeros
def make_resampled_transformation_vector(nii_obj,resample_dim=[4,4,4],standard_mask=True):
resamp_nii = make_resampled_transformation(nii_obj,resample_dim,standard_mask)
if standard_mask:
standard = get_standard_mask(voxdim=resample_dim[0])
return resamp_nii.get_data()[standard.get_data()!=0]
else:
return resamp_nii.get_data().flatten()
def test_unmasked_transformation():
image1 = nibabel.load(get_pair_images()[0])
nonzero_voxels = len(image1.get_data().flatten())
image1_vector = make_resampled_transformation_vector(image1,resample_dim=[2,2,2],standard_mask=False)
assert_equal(nonzero_voxels,len(image1_vector))
brain_4mm = get_standard_mask(4)
nonzero_voxels = len(brain_4mm.get_data().flatten())
image1_vector = make_resampled_transformation_vector(image1,resample_dim=[4,4,4],standard_mask=False)
assert_equal(nonzero_voxels,len(image1_vector))
def test_unmasked_transformation():
image1 = nibabel.load(get_pair_images()[0])
nonzero_voxels = len(image1.get_data().flatten())
image1_vector = make_resampled_transformation_vector(
image1, resample_dim=[2, 2, 2], standard_mask=False)
assert_equal(nonzero_voxels, len(image1_vector))
brain_4mm = get_standard_mask(4)
nonzero_voxels = len(brain_4mm.get_data().flatten())
image1_vector = make_resampled_transformation_vector(
image1, resample_dim=[4, 4, 4], standard_mask=False)
assert_equal(nonzero_voxels, len(image1_vector))
download_dir = "/ahba_data"
os.makedirs(download_dir)
for i, url in enumerate(urls):
print "Downloading %s" % url
urllib.urlretrieve(url, os.path.join(download_dir, "donor%d.zip" % (i + 1)))
zipfile.ZipFile(os.path.join(download_dir, "donor%d.zip" % (i + 1)))
# Dowloading MNI coordinates
urllib.urlretrieve(
"https://raw.githubusercontent.com/chrisfilo/alleninf/master/alleninf/data/corrected_mni_coordinates.csv",
os.path.join(download_dir, "corrected_mni_coordinates.csv"))
samples = pd.read_csv(os.path.join(download_dir, "corrected_mni_coordinates.csv"), index_col=0)
mni = get_standard_mask(voxdim=4)
reduced_coord = []
for coord_mni in samples[['corrected_mni_x', 'corrected_mni_y', 'corrected_mni_z']].values:
sample_counts = np.zeros(mni.shape, dtype=int)
coord_data = [int(round(i)) for i in nb.affines.apply_affine(npl.inv(mni.get_affine()), coord_mni)]
sample_counts[coord_data[0],
coord_data[1],
coord_data[2]] = 1
out_vector = sample_counts[mni.get_data()!=0]
idx = out_vector.argmax()
if idx == (out_vector == 1.0).sum() == 0:
idx = np.nan
reduced_coord.append(idx)
base = sys.argv[1]
data = "%s/data" % base
node_folder = "%s/groups" % data
results = "%s/results" % base # any kind of tsv/result file
decode_folder = "%s/decode" % base
if not os.path.exists(decode_folder):
os.mkdir(decode_folder)
# Images by Concepts data frame
labels_tsv = "%s/images_contrasts_df.tsv" % results
images = pandas.read_csv(labels_tsv, sep="\t", index_col=0)
output_folder = "%s/classification_final" % results
# Get standard mask, 4mm
standard_mask = get_standard_mask(4)
# Get all cognitive atlas concepts
all_concepts = get_concept().json
concepts = dict()
for concept in all_concepts:
concepts[concept["id"]] = str(concept["name"])
# You will need to copy abstracts.txt into this folder from the repo
abstracts = pandas.read_csv("%s/abstracts.txt" % decode_folder,
sep="\t",
index_col=0,
header=None)
abstracts.columns = ["text"]
# Here we can produce a similarity matrix to compare images A and B
import pandas
import nibabel
from pybraincompare.mr.datasets import get_standard_mask
from nilearn.masking import apply_mask
from scipi.stats import pearsonr
standard = get_standard_mask()
from glob import glob
amaps = glob("*groupA*.nii.gz")
bmaps = glob("*groupB*.nii.gz")
# Compare all A maps vs A maps
asim = pandas.DataFrame(index=amaps,columns=amaps)
bsim = pandas.DataFrame(index=bmaps,columns=bmaps)
for i in range(0,len(amaps)):
print "Processing %s of %s" %(i,len(amaps))
for j in range(0,len(amaps)):
amapi = nibabel.load(amaps[i])
amapj = nibabel.load(amaps[j])
bmapi = nibabel.load(bmaps[i])
bmapj = nibabel.load(bmaps[j])
vectorai = apply_mask(amapi,standard)
vectoraj = apply_mask(amapj,standard)
vectorbi = apply_mask(bmapi,standard)
vectorbj = apply_mask(bmapj,standard)
asim.loc[amaps[i],amaps[j]] = pearsonr(vectorai,vectoraj)[0]
labels_tsv = sys.argv[3]
image_lookup = sys.argv[4]
# image_pairs should be string of image pairs 1|2,3|4 for about 2000 (half) the image pairs
# We will calculate an accuracy across the image set, and build a null distribution by doing
# this procedure many times
image_pairs = image_pairs.split(",")
total = 0
correct = 0
# We will save a vector of
# Images by Concept data frame, our X
X = pandas.read_csv(labels_tsv,sep="\t",index_col=0)
# Get standard mask, 4mm
standard_mask=get_standard_mask(4)
# Dictionary to look up image files (4mm)
lookup = pickle.load(open(image_lookup,"rb"))
concepts = X.columns.tolist()
# We will go through each voxel (column) in a data frame of image data
image_paths = lookup.values()
mr = get_images_df(file_paths=image_paths,mask=standard_mask)
image_ids = [int(os.path.basename(x).split(".")[0]) for x in image_paths]
mr.index = image_ids
# We will go through each voxel (column) in a data frame of image data
mr = get_images_df(file_paths=group["in"] + group["out"],mask=standard_mask)
labels_tsv = sys.argv[3]
image_lookup = sys.argv[4]
# image_pairs should be string of image pairs 1|2,3|4 for about 2000 (half) the image pairs
# We will calculate an accuracy across the image set, and build a null distribution by doing
# this procedure many times
image_pairs = image_pairs.split(",")
total = 0
correct = 0
# We will save a vector of
# Images by Concept data frame, our X
X = pandas.read_csv(labels_tsv, sep="\t", index_col=0)
# Get standard mask, 4mm
standard_mask = get_standard_mask(4)
# Dictionary to look up image files (4mm)
lookup = pickle.load(open(image_lookup, "rb"))
concepts = X.columns.tolist()
# We will go through each voxel (column) in a data frame of image data
image_paths = lookup.values()
mr = get_images_df(file_paths=image_paths, mask=standard_mask)
image_ids = [int(os.path.basename(x).split(".")[0]) for x in image_paths]
mr.index = image_ids
# We will go through each voxel (column) in a data frame of image data
mr = get_images_df(file_paths=group["in"] + group["out"], mask=standard_mask)
image_paths = group["in"] + group["out"]
for i, url in enumerate(urls):
print "Downloading %s" % url
urllib.urlretrieve(url, os.path.join(download_dir,
"donor%d.zip" % (i + 1)))
zipfile.ZipFile(os.path.join(download_dir, "donor%d.zip" % (i + 1)))
# Dowloading MNI coordinates
urllib.urlretrieve(
"https://raw.githubusercontent.com/chrisfilo/alleninf/master/alleninf/data/corrected_mni_coordinates.csv",
os.path.join(download_dir, "corrected_mni_coordinates.csv"))
samples = pd.read_csv(os.path.join(download_dir,
"corrected_mni_coordinates.csv"),
index_col=0)
mni = get_standard_mask(voxdim=4)
reduced_coord = []
for coord_mni in samples[[
'corrected_mni_x', 'corrected_mni_y', 'corrected_mni_z'
]].values:
sample_counts = np.zeros(mni.shape, dtype=int)
coord_data = [
int(round(i))
for i in nb.affines.apply_affine(npl.inv(mni.get_affine()), coord_mni)
]
sample_counts[coord_data[0], coord_data[1], coord_data[2]] = 1
out_vector = sample_counts[mni.get_data() != 0]
idx = out_vector.argmax()
if idx == (out_vector == 1.0).sum() == 0:
base = sys.argv[1]
data = "%s/data" %base
node_folder = "%s/groups" %data
results = "%s/results" %base # any kind of tsv/result file
decode_folder = "%s/decode" %base
if not os.path.exists(decode_folder):
os.mkdir(decode_folder)
# Images by Concepts data frame
labels_tsv = "%s/images_contrasts_df.tsv" %results
images = pandas.read_csv(labels_tsv,sep="\t",index_col=0)
output_folder = "%s/classification_final" %results
# Get standard mask, 4mm
standard_mask=get_standard_mask(4)
# Get all cognitive atlas concepts
all_concepts = get_concept().json
concepts = dict()
for concept in all_concepts:
concepts[concept["id"]] = str(concept["name"])
# You will need to copy abstracts.txt into this folder from the repo
abstracts = pandas.read_csv("%s/abstracts.txt" %decode_folder,sep="\t",index_col=0,header=None)
abstracts.columns = ["text"]
# Functions to parse text
def remove_nonenglish_chars(text):
return re.sub("[^a-zA-Z]", " ", text)
base = sys.argv[1]
data = "%s/data" %base
node_folder = "%s/groups" %data
results = "%s/results" %base # any kind of tsv/result file
decode_folder = "%s/decode" %base
if not os.path.exists(decode_folder):
os.mkdir(decode_folder)
# Images by Concepts data frame
labels_tsv = "%s/images_contrasts_df.tsv" %results
images = pandas.read_csv(labels_tsv,sep="\t",index_col=0)
output_folder = "%s/classification_final" %results
# Get standard mask, 4mm
standard_mask=get_standard_mask(4)
# Load the regression params data frame
result = pickle.load(open("%s/regression_params_dfs.pkl" %output_folder,"rb"))
all_concepts = get_concept().json
concepts = dict()
for concept in all_concepts:
concepts[concept["id"]] = str(concept["name"])
# You will need to copy abstracts.txt into this folder from the repo
abstracts = pandas.read_csv("%s/abstracts.txt" %decode_folder,sep="\t",index_col=0,header=None)
abstracts.columns = ["text"]
# Functions to parse text
def remove_nonenglish_chars(text):
