def SkeletonizeSequentially(meta_filename):
# read in the data for this block
data = ReadMetaData(meta_filename)
# users must provide an output directory
assert (not data.SkeletonOutputDirectory() == None)
os.makedirs(data.SkeletonOutputDirectory(), exist_ok=True)
# compute the first step to save the walls of each file
for iz in range(data.StartZ(), data.EndZ()):
for iy in range(data.StartY(), data.EndY()):
for ix in range(data.StartX(), data.EndX()):
SaveAnchorWalls(data, iz, iy, ix)
# compute the second step to find the anchors between blocks
for iz in range(data.StartZ(), data.EndZ()):
for iy in range(data.StartY(), data.EndY()):
for ix in range(data.StartX(), data.EndX()):
ComputeAnchorPoints(data, iz, iy, ix)
# compute the third step to thin each block independently
for iz in range(data.StartZ(), data.EndZ()):
for iy in range(data.StartY(), data.EndY()):
for ix in range(data.StartX(), data.EndX()):
TopologicalThinning(data, iz, iy, ix)
# compute the fourth step to refine the skeleton
for label in range(1, data.NLabels()):
RefineSkeleton(data, label)
def CalculateBlockStatisticsSequentially(meta_filename):
data = ReadMetaData(meta_filename)
# iterate over all blocks
for iz in range(data.StartZ(), data.EndZ()):
for iy in range(data.StartY(), data.EndY()):
for ix in range(data.StartX(), data.EndX()):
CalculatePerBlockStatistics(data, iz, iy, ix)
CombineStatistics(data)
def CollectSurfacesSequentially(meta_filename):
# read in the data for this block
data = ReadMetaData(meta_filename)
assert (not data.SurfacesDirectory() == None)
# compute the first step to save the walls of each file
for iz in range(data.StartZ(), data.EndZ()):
for iy in range(data.StartY(), data.EndY()):
for ix in range(data.StartX(), data.EndX()):
GenerateSurfacesPerBlock(data, iz, iy, ix)
CombineSurfaceVoxels(data)
def CalculateSomataStatistics(meta_filename):
data = ReadMetaData(meta_filename)
somata_statistics = {}
# iterate over all blocks
for iz in range(data.StartZ(), data.EndZ()):
for iy in range(data.StartY(), data.EndY()):
for ix in range(data.StartX(), data.EndX()):
print('{} {:04d}z-{:04d}y-{:04d}x'.format(
meta_filename, iz, iy, ix))
# some datasets have no somata (default value)
upsampled_non_zero_voxels = 0
if data.SomataDownsampleRate():
somata = data.ReadSomataBlock(iz, iy, ix)
# get the number of non zero voxels
non_zero_voxels = np.count_nonzero(somata)
# the upsample factor is the number of voxels at full resolution correspond
# to one voxel at the downsampled resolution
upsample_factor = data.SomataDownsampleRate()**3
# the number of voxels masked as full resolution
upsampled_non_zero_voxels = upsample_factor * non_zero_voxels
somata_statistics[(iz, iy, ix)] = upsampled_non_zero_voxels
statistics_directory = '{}/statistics'.format(data.TempDirectory())
if not os.path.exists(statistics_directory):
os.makedirs(statistics_directory, exist_ok=True)
statistics_filename = '{}/somata-statistics.pickle'.format(
statistics_directory)
PickleData(somata_statistics, statistics_filename)
def EvaluateNeuralReconstructionIntegrity(meta_filename):
data = ReadMetaData(meta_filename)
synapses_per_label = {}
# read in all of the synapses from all of the blocks
synapse_directory = data.SynapseDirectory()
for iz in range(data.StartZ(), data.EndZ()):
for iy in range(data.StartY(), data.EndY()):
for ix in range(data.StartX(), data.EndX()):
synapses_filename = '{}/{:04d}z-{:04d}y-{:04d}x.pts'.format(synapse_directory, iz, iy, ix)
# ignore the local coordinates
block_synapses, _ = ReadPtsFile(data, synapses_filename)
# add all synapses for each label in this block to the global synapses
for label in block_synapses.keys():
if not label in synapses_per_label:
synapses_per_label[label] = []
synapses_per_label[label] += block_synapses[label]
# get the output filename
evaluation_directory = data.EvaluationDirectory()
if not os.path.exists(evaluation_directory):
os.makedirs(evaluation_directory, exist_ok=True)
output_filename = '{}/nri-results.txt'.format(evaluation_directory)
fd = open(output_filename, 'w')
# keep track of global statistics
total_true_positives = 0
total_false_positives = 0
total_false_negatives = 0
# for each label, find if synapses correspond to endpoints
for label in range(1, data.NLabels()):
# read the refined skeleton for this synapse
skeleton_directory = '{}/skeletons'.format(data.SkeletonOutputDirectory())
skeleton_filename = '{}/{:016d}.pts'.format(skeleton_directory, label)
# skip over labels not processed
if not os.path.exists(skeleton_filename): continue
# get the synapses only for this one label
synapses = synapses_per_label[label]
# ignore the local coordinates
skeletons, _ = ReadPtsFile(data, skeleton_filename)
skeleton = skeletons[label]
# read in the somata surfaces (points on the surface should not count as endpoints)
somata_directory = '{}/somata_surfaces'.format(data.TempDirectory())
somata_filename = '{}/{:016d}.pts'.format(somata_directory, label)
# path may not exist if soma not found
if os.path.exists(somata_filename):
somata_surfaces, _ = ReadPtsFile(data, somata_filename)
somata_surface = set(somata_surfaces[label])
else:
somata_surface = set()
# get the endpoints in this skeleton for this label
endpoints = FindEndpoints(data, skeleton, somata_surface)
true_positives, false_positives, false_negatives = CalculateNeuralReconstructionIntegrityScore(data, synapses, endpoints)
# if there are no true positives the NRI score is 0
if true_positives == 0:
nri_score = 0
else:
precision = true_positives / float(true_positives + false_positives)
recall = true_positives / float(true_positives + false_negatives)
nri_score = 2 * (precision * recall) / (precision + recall)
print ('Label: {}'.format(label))
print (' True Positives: {:10d}'.format(true_positives))
print (' False Positives: {:10d}'.format(false_positives))
print (' False Negatives: {:10d}'.format(false_negatives))
print (' NRI Score: {:0.4f}'.format(nri_score))
fd.write ('Label: {}\n'.format(label))
fd.write (' True Positives: {:10d}\n'.format(true_positives))
fd.write (' False Positives: {:10d}\n'.format(false_positives))
fd.write (' False Negatives: {:10d}\n'.format(false_negatives))
fd.write (' NRI Score: {:0.4f}\n'.format(nri_score))
# update the global stats
total_true_positives += true_positives
total_false_positives += false_positives
total_false_negatives += false_negatives
precision = total_true_positives / float(total_true_positives + total_false_positives)
recall = total_true_positives / float(total_true_positives + total_false_negatives)
nri_score = 2 * (precision * recall) / (precision + recall)
print ('Total Volume'.format(label))
print (' True Positives: {:10d}'.format(total_true_positives))
print (' False Positives: {:10d}'.format(total_false_positives))
print (' False Negatives: {:10d}'.format(total_false_negatives))
print (' NRI Score: {:0.4f}'.format(nri_score))
fd.write ('Total Volume\n'.format(label))
fd.write (' True Positives: {:10d}\n'.format(total_true_positives))
fd.write (' False Positives: {:10d}\n'.format(total_false_positives))
fd.write (' False Negatives: {:10d}\n'.format(total_false_negatives))
fd.write (' NRI Score: {:0.4f}\n'.format(nri_score))
def EvaluateHoleFilling(meta_filename):
data = ReadMetaData(meta_filename)
# make sure a results folder is specified
assert (not data.EvaluationDirectory() == None)
hole_sizes = {}
neighbor_label_dicts = {}
# read in the hole sizes from each block
for iz in range(data.StartZ(), data.EndZ()):
for iy in range(data.StartY(), data.EndY()):
for ix in range(data.StartX(), data.EndX()):
tmp_block_directory = data.TempBlockDirectory(iz, iy, ix)
# read the saved hole sizes for this block
hole_sizes_filename = '{}/hole-sizes.pickle'.format(tmp_block_directory)
holes_sizes_per_block = ReadPickledData(hole_sizes_filename)
for label in holes_sizes_per_block:
hole_sizes[label] = holes_sizes_per_block[label]
# any value already determined in the local step mush have no neighbors
associated_label_dict = ReadPickledData('{}/associated-label-set-local.pickle'.format(tmp_block_directory))
for label in associated_label_dict:
neighbor_label_dicts[label] = []
# read in the neighbor label dictionary that maps values to its neighbors
tmp_directory = data.TempDirectory()
neighbor_label_filename = '{}/hole-filling-neighbor-label-dict-global.pickle'.format(tmp_directory)
neighbor_label_dict_global = ReadPickledData(neighbor_label_filename)
associated_label_filename = '{}/hole-filling-associated-labels.pickle'.format(tmp_directory)
associated_label_dict = ReadPickledData(associated_label_filename)
# make sure that the keys are identical for hole sizes and associated labels (sanity check)
assert (sorted(hole_sizes.keys()) == sorted(associated_label_dict.keys()))
# make sure no query component in the global dictionary occurs in the local dictionary
for label in neighbor_label_dict_global.keys():
assert (not label in neighbor_label_dicts)
# create a unified neighbor labels dictionary that combines local and global information
neighbor_label_dicts.update(neighbor_label_dict_global)
# make sure that the keys are identical for hole sizes and the neighbor label dicts
assert (sorted(hole_sizes.keys()) == sorted(neighbor_label_dicts.keys()))
# union find data structure to link together holes across blocks
class UnionFindElement:
def __init__(self, label):
self.label = label
self.parent = self
self.rank = 0
def Find(element):
if not element.parent == element:
element.parent = Find(element.parent)
return element.parent
def Union(element_one, element_two):
root_one = Find(element_one)
root_two = Find(element_two)
if root_one == root_two: return
if root_one.rank < root_two.rank:
root_one.parent = root_two
elif root_one.rank > root_two.rank:
root_two.parent = root_one
else:
root_two.parent = root_one
root_one.rank = root_one.rank + 1
union_find_elements = {}
for label in neighbor_label_dicts.keys():
# skip over elements that remain background
if not associated_label_dict[label]: continue
union_find_elements[label] = UnionFindElement(label)
for label in neighbor_label_dicts.keys():
# skip over elements that remain background
if not associated_label_dict[label]: continue
for neighbor_label in neighbor_label_dicts[label]:
# skip over the actual neuron label
if neighbor_label > 0: continue
# merge these two labels together
Union(union_find_elements[label], union_find_elements[neighbor_label])
root_holes_sizes = {}
# go through all labels in the union find data structure and update the hole size for the parent
for label in union_find_elements.keys():
root_label = Find(union_find_elements[label])
# create this hole if it does not already exist
if not root_label.label in root_holes_sizes:
root_holes_sizes[root_label.label] = 0
root_holes_sizes[root_label.label] += hole_sizes[label]
# read in the statistics data to find total volume size
statistics_directory = '{}/statistics'.format(data.TempDirectory())
statistics_filename = '{}/combined-statistics.pickle'.format(statistics_directory)
volume_statistics = ReadPickledData(statistics_filename)
total_volume = volume_statistics['neuronal_volume']
holes = []
small_holes = 0
for root_label in root_holes_sizes.keys():
if root_holes_sizes[root_label] < 5:
small_holes += 1
holes.append(root_holes_sizes[root_label])
# get statistics on the number of holes
nholes = len(holes)
total_hole_volume = sum(holes)
print ('Percent Small: {}'.format(100.0 * small_holes / nholes))
print ('No. Holes: {}'.format(nholes))
print ('Total Volume: {} ({:0.2f}%)'.format(total_hole_volume, 100.0 * total_hole_volume / total_volume))
def MapSynapsesAndSurfaces(input_meta_filename, output_meta_filename):
# read in the meta data files
input_data = ReadMetaData(input_meta_filename)
output_data = ReadMetaData(output_meta_filename)
input_synapse_directory = input_data.SynapseDirectory()
output_synapse_directory = output_data.SynapseDirectory()
# create the output synapse directory if it does not exist
if not os.path.exists(output_synapse_directory):
os.makedirs(output_synapse_directory, exist_ok=True)
# read in all of the input syanpses
input_synapses = {}
for iz in range(input_data.StartZ(), input_data.EndZ()):
for iy in range(input_data.StartY(), input_data.EndY()):
for ix in range(input_data.StartX(), input_data.EndX()):
input_synapse_filename = '{}/{:04d}z-{:04d}y-{:04d}x.pts'.format(input_synapse_directory, iz, iy, ix)
# ignore the local points
global_pts, _ = ReadPtsFile(input_data, input_synapse_filename)
for label in global_pts:
if not label in input_synapses:
input_synapses[label] = []
# add the array from this block to the input synapses
input_synapses[label] += global_pts[label]
# create an output dictionary of synapses per block
output_synapses_per_block = {}
# iterate over all output blocks
for iz in range(output_data.StartZ(), output_data.EndZ()):
for iy in range(output_data.StartY(), output_data.EndY()):
for ix in range(output_data.StartX(), output_data.EndX()):
# create empty synapses_per_block dictionaries whose keys will be labels
output_synapses_per_block[(iz, iy, ix)] = {}
# for every label, go through the discovered synapses from the input data
for label in input_synapses.keys():
input_synapses_per_label = input_synapses[label]
for input_global_index in input_synapses_per_label:
# the global iz, iy, ix coordinates remain the same across blocks
global_iz, global_iy, global_ix = input_data.GlobalIndexToIndices(input_global_index)
# get the new block from the global coordinates
output_block_iz = global_iz // output_data.BlockZLength()
output_block_iy = global_iy // output_data.BlockYLength()
output_block_ix = global_ix // output_data.BlockXLength()
if not label in output_synapses_per_block[(output_block_iz, output_block_iy, output_block_ix)]:
output_synapses_per_block[(output_block_iz, output_block_iy, output_block_ix)][label] = []
# get the new global index
output_global_index = output_data.GlobalIndicesToIndex(global_iz, global_iy, global_ix)
output_synapses_per_block[(output_block_iz, output_block_iy, output_block_ix)][label].append(output_global_index)
# write all of the synapse block files
output_synapses_directory = output_data.SynapseDirectory()
for iz in range(output_data.StartZ(), output_data.EndZ()):
for iy in range(output_data.StartY(), output_data.EndY()):
for ix in range(output_data.StartX(), output_data.EndX()):
output_synapse_filename = '{}/{:04d}z-{:04d}y-{:04d}x.pts'.format(output_synapse_directory, iz, iy, ix)
# write the pts file (use global indices)
WritePtsFile(output_data, output_synapse_filename, output_synapses_per_block[(iz, iy, ix)], input_local_indices = False)
# get the input/output directories for the surfaces
input_surfaces_directory = input_data.SurfacesDirectory()
output_surfaces_directory = output_data.SurfacesDirectory()
# create the output synapse directory if it does not exist
if not os.path.exists(output_surfaces_directory):
os.makedirs(output_surfaces_directory, exist_ok=True)
# iterate over all labels
for label in range(1, input_data.NLabels()):
start_time = time.time()
# skip over labels that do not exist
input_surface_filename = '{}/{:016d}.pts'.format(input_surfaces_directory, label)
if not os.path.exists(input_surface_filename): continue
# read in the input global points
input_global_points, _ = ReadPtsFile(input_data, input_surface_filename)
# create an empty dictionary for the output points
output_global_points = {}
output_global_points[label] = []
for input_global_index in input_global_points[label]:
# the global iz, iy, ix coordinates remain the same across blocks
global_iz, global_iy, global_ix = input_data.GlobalIndexToIndices(input_global_index)
# get the new global index
output_global_index = output_data.GlobalIndicesToIndex(global_iz, global_iy, global_ix)
output_global_points[label].append(output_global_index)
# write the new surface filename
output_surface_filename = '{}/{:016d}.pts'.format(output_surfaces_directory, label)
WritePtsFile(output_data, output_surface_filename, output_global_points, input_local_indices = False)
print ('Completed label {} in {:0.2f} seconds'.format(label, time.time() - start_time))
def ConvertSynapsesAndProject(meta_filename, input_synapse_directory, xyz,
conversion_rate):
data = ReadMetaData(meta_filename)
resolution = data.Resolution()
# create an empty set of synapses
synapses = {}
# iterate over all labels
for label in range(1, data.NLabels()):
# read the surfaces for this label
surface_filename = '{}/{:016d}.pts'.format(data.SurfacesDirectory(),
label)
# some surfaces (i.e., labels) will not exist in the volume
if not os.path.exists(surface_filename): continue
# read in the surface points, ignore the local coordinates
surfaces, _ = ReadPtsFile(data, surface_filename)
surface = surfaces[label]
npts = len(surface)
surface_point_cloud = np.zeros((npts, 3), dtype=np.int32)
for index, iv in enumerate(surface):
iz, iy, ix = data.GlobalIndexToIndices(iv)
surface_point_cloud[index, :] = (iz * resolution[OR_Z],
iy * resolution[OR_Y],
ix * resolution[OR_X])
# create an empty array for the synapses
synapses[label] = []
projected = 0
missed = 0
# read in the original synapses
input_synapse_filename = '{}/syn_{:04}.txt'.format(
input_synapse_directory, label)
if os.path.exists(input_synapse_filename):
with open(input_synapse_filename, 'r') as fd:
for line in fd:
# separate the line into coordinates
coordinates = line.strip().split()
if xyz:
ix = round(int(coordinates[0]) / conversion_rate[OR_X])
iy = round(int(coordinates[1]) / conversion_rate[OR_Y])
iz = round(int(coordinates[2]) / conversion_rate[OR_Z])
else:
iz = round(int(coordinates[0]) / conversion_rate[OR_Z])
iy = round(int(coordinates[1]) / conversion_rate[OR_Y])
ix = round(int(coordinates[2]) / conversion_rate[OR_X])
# create a 2D vector for this point
vec = np.zeros((1, 3), dtype=np.int32)
vec[0, :] = (iz * resolution[OR_Z], iy * resolution[OR_Y],
ix * resolution[OR_X])
closest_point = surface[scipy.spatial.distance.cdist(
surface_point_cloud, vec).argmin()]
closest_iz, closest_iy, closest_ix = data.GlobalIndexToIndices(
closest_point)
deltaz = resolution[OR_Z] * (iz - closest_iz)
deltay = resolution[OR_Y] * (iy - closest_iy)
deltax = resolution[OR_X] * (ix - closest_ix)
distance = math.sqrt(deltaz * deltaz + deltay * deltay +
deltax * deltax)
# skip distances that are clearly off (over 200 nanometers)
max_deviation = 800
if distance < max_deviation:
# add to the list of valid synapses
synapses[label].append(closest_point)
projected += 1
else:
missed += 1
print('Synapses within {} nanometers from surface: {}'.format(
max_deviation, projected))
print('Synapses over {} nanometers from surface: {}'.format(
max_deviation, missed))
# divide all synapses into blocks
synapses_per_block = {}
# iterate over all blocks
for iz in range(data.StartZ(), data.EndZ()):
for iy in range(data.StartY(), data.EndY()):
for ix in range(data.StartX(), data.EndX()):
# create empty synapses_per_block dictionaries whose keys will be labels
synapses_per_block[(iz, iy, ix)] = {}
# for every label, iterate over the discovered synapses
for label in synapses.keys():
synapses_per_label = synapses[label]
# iterate over all of the projected synapses
for global_index in synapses_per_label:
global_iz, global_iy, global_ix = data.GlobalIndexToIndices(
global_index)
block_iz = global_iz // data.BlockZLength()
block_iy = global_iy // data.BlockYLength()
block_ix = global_ix // data.BlockXLength()
# create the array for this label per block if it does not exist
if not label in synapses_per_block[(block_iz, block_iy, block_ix)]:
synapses_per_block[(block_iz, block_iy, block_ix)][label] = []
synapses_per_block[(block_iz, block_iy,
block_ix)][label].append(global_index)
# write all of the synapse block files
synapse_directory = data.SynapseDirectory()
if not os.path.exists(synapse_directory):
os.makedirs(synapse_directory, exist_ok=True)
for iz in range(data.StartZ(), data.EndZ()):
for iy in range(data.StartY(), data.EndY()):
for ix in range(data.StartX(), data.EndX()):
synapse_filename = '{}/{:04d}z-{:04d}y-{:04d}x.pts'.format(
synapse_directory, iz, iy, ix)
# write the pts file (use global indices)
WritePtsFile(data,
synapse_filename,
synapses_per_block[(iz, iy, ix)],
input_local_indices=False)
from blockbased_synapseaware.benchmark.kimimaro.connectskeletons import ConnectSkeletons
# pass arguments
# read passed arguments
if len(sys.argv) != 2:
raise ValueError(" Scripts needs exactley 1 input arguments (Prefix) ")
else:
meta_fp = sys.argv[1]
# read in the data for this block
data = ReadMetaData(meta_fp)
# Redirect stdout and stderr
RedirectOutStreams(data.BlockSize(), "KM", 2, "all", "all", "all")
# check that beforehand step has executed successfully
for iz in range(data.StartZ(), data.EndZ()):
for iy in range(data.StartY(), data.EndY()):
for ix in range(data.StartX(), data.EndX()):
CheckSuccessFile(data.BlockSize(), "KM", 1, iz, iy, ix)
# users must provide an output directory
assert (not data.SkeletonOutputDirectory() == None)
os.makedirs(data.SkeletonOutputDirectory(), exist_ok=True)
# compute the second step to find adjacencies between borders
ConnectSkeletons(data)
# Create and Write Success File
WriteSuccessFile(data.BlockSize(), "KM", 2, "all", "all", "all")