def process(self, dataset, i, k):
print 'Register Data %s with ICP(centerline)...(Type %d)' % (
self.ini.file.name_result[i], k)
dataset['fix'].pointSet.data['Centerline'] = calCenterlineFromContour(
dataset['fix'].pointSet.data, k, dataset['fix'].getData())
dataset['mov'].pointSet.data['Centerline'] = calCenterlineFromContour(
dataset['mov'].pointSet.data, k, dataset['mov'].getData())
data, point, para = self.icp.register(dataset['fix'], dataset['mov'],
1)
resultData = db.ResultData(data,
db.ImageInfo(dataset['fix'].info.data),
point)
resultData.info.addData('fix', 1)
resultData.info.addData('move', 2)
resultData.info.addData('transform', para)
print 'Done!'
mean_dis, mean_whole, max_dis, max_whole = self.surfaceerror.analysis(
resultData, dataset['fix'].getPointSet('Contour').copy(),
dataset['mov'].getPointSet('Contour').copy(),
dataset['mov'].getPointSet('Mask').copy(),
dataset['mov'].getResolution().tolist())
print 'Contour Error Done! Whole mean is %0.2fmm.' % mean_whole
for j in range(3):
self.sheet.write(j + 1 + 4 * k, i + 2, float(mean_dis[j]))
self.sheet.write(4 + 4 * k, i + 2, mean_whole)
self.book.save('./Result/Center_merge.xls')
del data, point, resultData
def load(self, i):
dataset = {'mov': [], 'fix': []}
data, info, point = db.loadMatData(self.path + self.ini.file.datadir + '/Contour/'
+ self.ini.file.name_mov[i] + '.mat', None)
point['Centerline'] = calCenterlineFromContour(point)
dataset['mov'] = db.BasicData(data, info, point)
data, info, point = db.loadMatData(self.path + self.ini.file.datadir + '/Contour/'
+ self.ini.file.name_fix[i] + '.mat', None)
point['Centerline'] = calCenterlineFromContour(point)
dataset['fix'] = db.BasicData(data, info, point)
print 'Data %s loaded!' % self.ini.file.name_result[i]
return dataset
def load(self, k, i):
dataset = {'mov': [], 'fix': []}
data, info, point = db.loadMatData(self.path + self.ini.file.datadir + '/Contour/'
+ self.ini.file.name_fix[i] + '.mat', None)
point['Centerline'] = calCenterlineFromContour(point)
fileData = db.BasicData(data, info, point)
dataset['fix'] = fileData
data, info, point = db.loadMatData(self.path + self.ini.file.datadir + '/Contour/'
+ self.ini.file.name_mov[i] + '.mat', None)
point['Centerline'] = calCenterlineFromContour(point)
fileData = db.BasicData(data, info, point)
dataset['mov'] = fileData
print 'Data %s loaded!' % self.ini.file.name_result[i]
return dataset
def process(self, dataset, i, k):
print 'Register Data %s with ICP(centerline)...(Type %d)' % (self.ini.file.name_result[i], k)
dataset['fix'].pointSet.data['Centerline'] = calCenterlineFromContour(dataset['fix'].pointSet.data, k, dataset['fix'].getData())
dataset['mov'].pointSet.data['Centerline'] = calCenterlineFromContour(dataset['mov'].pointSet.data, k, dataset['mov'].getData())
data, point, para = self.icp.register(dataset['fix'], dataset['mov'], 1)
resultData = db.ResultData(data, db.ImageInfo(dataset['fix'].info.data), point)
resultData.info.addData('fix', 1)
resultData.info.addData('move', 2)
resultData.info.addData('transform', para)
print 'Done!'
mean_dis, mean_whole, max_dis, max_whole = self.surfaceerror.analysis(resultData, dataset['fix'].getPointSet('Contour').copy(), dataset['mov'].getPointSet('Contour').copy(), dataset['mov'].getPointSet('Mask').copy(), dataset['mov'].getResolution().tolist())
print 'Contour Error Done! Whole mean is %0.2fmm.' % mean_whole
for j in range(3):
self.sheet.write(j + 1 + 4 * k, i + 2, float(mean_dis[j]))
self.sheet.write(4 + 4 * k, i + 2, mean_whole)
self.book.save('./Result/Center_merge.xls')
del data, point, resultData
def process(self, dataset, i):
def autoDetectContour(point, cnt, start, end, delta, res, type):
self.new_points = npy.append(self.new_points, point, 0)
points = point[:, :-2]
d = 20
count = 0
for i in range(start + delta, end + delta, delta):
center = calCentroidFromContour(points).reshape(2)
image = dataset[type].getData()[i, :, :].transpose().copy()
image = (image - npy.min(image)) / (npy.max(image) -
npy.min(image)) * 255
down = npy.max([npy.ceil(center[0] - d / res[0]), 0])
up = npy.min(
[npy.floor(center[0] + d / res[0]), image.shape[0]])
left = npy.max([npy.ceil(center[1] - d / res[1]), 0])
right = npy.min(
[npy.floor(center[1] + d / res[1]), image.shape[1]])
crop_image = image[down:up, left:right]
center -= [down, left]
result = ac_segmentation(center, crop_image)
a1 = ac_area(points.transpose(), image.shape)
a2 = ac_area(result, crop_image.shape)
rate = a2 * 1.0 / a1
if rate >= min(1.5 + count * 0.2, 2.1) or rate <= 0.7:
temp_array = points.copy()
if cnt != 1 and rate > 0.7:
count += 1
else:
temp_array = result.transpose().copy()
temp_array[:, :2] += [down, left]
count = 0
points = temp_array.copy()
temp_array = npy.insert(temp_array, 2, [[i], [cnt]], 1)
self.new_points = npy.append(self.new_points, temp_array, 0)
sys.stdout.write(str(i) + ',')
sys.stdout.flush()
print ' '
# Segmentation of data
print 'Segment Data %s...' % self.ini.file.name_result[i]
tmp = dataset['fix'].pointSet.data['Contour']
tmp = tmp[tmp[:, 0] >= 0]
self.new_points = npy.array([[-1, -1, -1, -1]], dtype=npy.float32)
bottom = int(npy.round(npy.min(tmp[:, 2])))
bif = int(db.getBifurcation(tmp) + 0.5)
up = int(npy.round(npy.max(tmp[:, 2])))
bottom += (bif - bottom) / 2
up -= (up - bif) / 2
point_vital = [0] * 3
point_vital[0] = tmp[(npy.round(tmp[:, -1]) == 0)
& (npy.round(tmp[:, 2]) == bottom)].copy()
point_vital[1] = tmp[(npy.round(tmp[:, -1]) == 1)
& (npy.round(tmp[:, 2]) == up)].copy()
point_vital[2] = tmp[(npy.round(tmp[:, -1]) == 2)
& (npy.round(tmp[:, 2]) == up)].copy()
autoDetectContour(point_vital[0], 0, bottom, bif, 1,
dataset['fix'].getResolution().tolist(), 'fix')
autoDetectContour(point_vital[1], 1, up, bif, -1,
dataset['fix'].getResolution().tolist(), 'fix')
autoDetectContour(point_vital[2], 2, up, bif, -1,
dataset['fix'].getResolution().tolist(), 'fix')
print ' '
print 'Finish segmentation for fix data. '
pointset = {'Contour': self.new_points}
dataset['fix'].pointSet.data['Centerline'] = calCenterlineFromContour(
pointset)
self.new_points_fix = self.new_points.copy()
# For mov data
tmp = dataset['mov'].pointSet.data['Contour']
tmp = tmp[tmp[:, 0] >= 0]
self.new_points = npy.array([[-1, -1, -1, -1]], dtype=npy.float32)
bottom = int(npy.round(npy.min(tmp[:, 2])))
bif = int(db.getBifurcation(tmp) + 0.5)
up = int(npy.round(npy.max(tmp[:, 2])))
bottom += (bif - bottom) / 2
up -= (up - bif) / 2
point_vital = [0] * 3
point_vital[0] = tmp[(npy.round(tmp[:, -1]) == 0)
& (npy.round(tmp[:, 2]) == bottom)].copy()
point_vital[1] = tmp[(npy.round(tmp[:, -1]) == 1)
& (npy.round(tmp[:, 2]) == up)].copy()
point_vital[2] = tmp[(npy.round(tmp[:, -1]) == 2)
& (npy.round(tmp[:, 2]) == up)].copy()
autoDetectContour(point_vital[0], 0, bottom, bif, 1,
dataset['mov'].getResolution().tolist(), 'mov')
autoDetectContour(point_vital[1], 1, up, bif, -1,
dataset['mov'].getResolution().tolist(), 'mov')
autoDetectContour(point_vital[2], 2, up, bif, -1,
dataset['mov'].getResolution().tolist(), 'mov')
print ' '
print 'Finish segmentation for mov data. '
pointset = {'Contour': self.new_points}
dataset['mov'].pointSet.data['Centerline'] = calCenterlineFromContour(
pointset)
self.new_points_mov = self.new_points.copy()
# ICP with centerline without label
print 'Register Data %s with ICP(centerline) without label...' % self.ini.file.name_result[
i]
data, point, para = self.icp.register(dataset['fix'],
dataset['mov'],
1,
op=True)
resultData = db.ResultData(data,
db.ImageInfo(dataset['fix'].info.data),
point)
resultData.info.addData('fix', 1)
resultData.info.addData('move', 2)
resultData.info.addData('transform', para)
print 'Done!'
mean_dis, mean_whole, max_dis, max_whole = self.surfaceerror.analysis(
resultData, dataset['fix'].getPointSet('Contour').copy(),
dataset['mov'].getPointSet('Contour').copy(),
dataset['mov'].getPointSet('Mask').copy(),
dataset['mov'].getResolution().tolist())
print 'Contour Error Done! Whole mean is %0.2fmm.' % mean_whole
dice_index, dice_index_all = self.areaerror.analysis(
resultData, dataset['fix'].getPointSet('Contour').copy())
print 'Area Error Done! Whole Dice index is %0.3f.' % dice_index_all
self.sheet2.write(0, i + 2, self.ini.file.name_result[i])
for j in range(3):
self.sheet2.write(j + 1, i + 2, mean_dis[j])
self.sheet2.write(j + 5, i + 2, max_dis[j])
self.sheet2.write(j + 9, i + 2, dice_index[j])
self.sheet2.write(4, i + 2, mean_whole)
self.sheet2.write(8, i + 2, max_whole)
self.sheet2.write(12, i + 2, dice_index_all)
self.book.save(self.path + self.ini.file.savedir +
'multicontrast_seg_feature.xls')
del data, point, resultData
# ICP with centerline with label
print 'Register Data %s with ICP(centerline) with label...' % self.ini.file.name_result[
i]
data, point, para = self.icp.register(dataset['fix'],
dataset['mov'],
1,
op=False)
resultData = db.ResultData(data,
db.ImageInfo(dataset['fix'].info.data),
point)
resultData.info.addData('fix', 1)
resultData.info.addData('move', 2)
resultData.info.addData('transform', para)
print 'Done!'
mean_dis, mean_whole, max_dis, max_whole = self.surfaceerror.analysis(
resultData, dataset['fix'].getPointSet('Contour').copy(),
dataset['mov'].getPointSet('Contour').copy(),
dataset['mov'].getPointSet('Mask').copy(),
dataset['mov'].getResolution().tolist())
print 'Contour Error Done! Whole mean is %0.2fmm.' % mean_whole
dice_index, dice_index_all = self.areaerror.analysis(
resultData, dataset['fix'].getPointSet('Contour').copy())
print 'Area Error Done! Whole Dice index is %0.3f.' % dice_index_all
self.sheet4.write(0, i + 2, self.ini.file.name_result[i])
for j in range(3):
self.sheet4.write(j + 1, i + 2, mean_dis[j])
self.sheet4.write(j + 5, i + 2, max_dis[j])
self.sheet4.write(j + 9, i + 2, dice_index[j])
self.sheet4.write(4, i + 2, mean_whole)
self.sheet4.write(8, i + 2, max_whole)
self.sheet4.write(12, i + 2, dice_index_all)
self.book.save(self.path + self.ini.file.savedir +
'multicontrast_seg_feature.xls')
del data, point, resultData
fix_points = dataset['fix'].getPointSet('Contour').copy()
dataset['fix'].pointSet.data['Contour'] = self.new_points_fix
mov_points = dataset['mov'].getPointSet('Contour').copy()
dataset['mov'].pointSet.data['Contour'] = self.new_points_mov
print 'Saving Data %s...' % self.ini.file.name_result[i]
db.saveMatData(
self.path + self.ini.file.savedir + self.ini.file.name_result[i] +
'_snap.mat', [dataset['fix']], 0)
db.saveMatData(
self.path + self.ini.file.savedir + self.ini.file.name_result[i] +
'_merge.mat', [dataset['mov']], 0)
print 'Done!'
# ICP with contour without label
print 'Register Data %s with ICP(contour) without label...' % self.ini.file.name_result[
i]
data, point, para = self.icp.register(dataset['fix'],
dataset['mov'],
0,
op=False)
resultData = db.ResultData(data,
db.ImageInfo(dataset['fix'].info.data),
point)
resultData.info.addData('fix', 1)
resultData.info.addData('move', 2)
resultData.info.addData('transform', para)
print 'Done!'
mean_dis, mean_whole, max_dis, max_whole = self.surfaceerror.analysis(
resultData, fix_points.copy(), mov_points.copy(),
dataset['mov'].getPointSet('Mask').copy(),
dataset['mov'].getResolution().tolist())
print 'Contour Error Done! Whole mean is %0.2fmm.' % mean_whole
para = resultData.info.getData('transform')
R = ml.mat(para[:9]).reshape(3, 3)
T = ml.mat(para[9:12]).T
T = R.I * T
T = -T
tmp_con, result_center_points = util.resliceTheResultPoints(
mov_points, None, 20, dataset['mov'].getResolution().tolist(),
dataset['fix'].getResolution().tolist(), False, R, T)
resultData.pointSet.data['Contour'] = tmp_con
dice_index, dice_index_all = self.areaerror.analysis(
resultData, fix_points.copy())
print 'Area Error Done! Whole Dice index is %0.3f.' % dice_index_all
self.sheet1.write(0, i + 2, self.ini.file.name_result[i])
for j in range(3):
self.sheet1.write(j + 1, i + 2, mean_dis[j])
self.sheet1.write(j + 5, i + 2, max_dis[j])
self.sheet1.write(j + 9, i + 2, dice_index[j])
self.sheet1.write(4, i + 2, mean_whole)
self.sheet1.write(8, i + 2, max_whole)
self.sheet1.write(12, i + 2, dice_index_all)
self.book.save(self.path + self.ini.file.savedir +
'multicontrast_seg_feature.xls')
del data, point, resultData
# ICP with contour with label
print 'Register Data %s with ICP(contour) with label...' % self.ini.file.name_result[
i]
data, point, para = self.icp.register(dataset['fix'],
dataset['mov'],
0,
op=True)
resultData = db.ResultData(data,
db.ImageInfo(dataset['fix'].info.data),
point)
resultData.info.addData('fix', 1)
resultData.info.addData('move', 2)
resultData.info.addData('transform', para)
print 'Done!'
mean_dis, mean_whole, max_dis, max_whole = self.surfaceerror.analysis(
resultData, fix_points.copy(), mov_points.copy(),
dataset['mov'].getPointSet('Mask').copy(),
dataset['mov'].getResolution().tolist())
print 'Contour Error Done! Whole mean is %0.2fmm.' % mean_whole
para = resultData.info.getData('transform')
R = ml.mat(para[:9]).reshape(3, 3)
T = ml.mat(para[9:12]).T
T = R.I * T
T = -T
tmp_con, result_center_points = util.resliceTheResultPoints(
mov_points, None, 20, dataset['mov'].getResolution().tolist(),
dataset['fix'].getResolution().tolist(), False, R, T)
resultData.pointSet.data['Contour'] = tmp_con
dice_index, dice_index_all = self.areaerror.analysis(
resultData, fix_points.copy())
print 'Area Error Done! Whole Dice index is %0.3f.' % dice_index_all
self.sheet3.write(0, i + 2, self.ini.file.name_result[i])
for j in range(3):
self.sheet3.write(j + 1, i + 2, mean_dis[j])
self.sheet3.write(j + 5, i + 2, max_dis[j])
self.sheet3.write(j + 9, i + 2, dice_index[j])
self.sheet3.write(4, i + 2, mean_whole)
self.sheet3.write(8, i + 2, max_whole)
self.sheet3.write(12, i + 2, dice_index_all)
self.book.save(self.path + self.ini.file.savedir +
'multicontrast_seg_feature.xls')
del data, point, resultData
del self.new_points, fix_points, self.new_points_fix, self.new_points_mov, mov_points, tmp_con, result_center_points
def process(self, dataset, i):
def autoDetectContour(point, cnt, start, end, delta, res):
self.new_points = npy.append(self.new_points, point, 0)
points = point[:, :-2]
d = 20
count = 0
for i in range(start + delta, end + delta, delta):
center = calCentroidFromContour(points).reshape(2)
image = dataset['fix'].getData()[i, :, :].transpose().copy()
image = (image - npy.min(image)) / (npy.max(image) - npy.min(image)) * 255
down = npy.max([npy.ceil(center[0] - d / res[0]), 0])
up = npy.min([npy.floor(center[0] + d / res[0]), image.shape[0]])
left = npy.max([npy.ceil(center[1] - d / res[1]), 0])
right = npy.min([npy.floor(center[1] + d / res[1]), image.shape[1]])
crop_image = image[down : up, left : right]
center -= [down, left]
result = ac_segmentation(center, crop_image)
a1 = ac_area(points.transpose(), image.shape)
a2 = ac_area(result, crop_image.shape)
rate = a2 * 1.0 / a1
if rate >= min(1.5 + count * 0.2, 2.1) or rate <= 0.7:
temp_array = points.copy()
if cnt != 1 and rate > 0.7:
count += 1
else:
temp_array = result.transpose().copy()
temp_array[:, :2] += [down, left]
count = 0
points = temp_array.copy()
temp_array = npy.insert(temp_array, 2, [[i], [cnt]], 1)
self.new_points = npy.append(self.new_points, temp_array, 0)
sys.stdout.write(str(i) + ',')
sys.stdout.flush()
print ' '
# Segmentation of data
print 'Segment Data %s...' % self.ini.file.name_result[i]
tmp = dataset['fix'].pointSet.data['Contour']
tmp = tmp[tmp[:, 0] >= 0]
self.new_points = npy.array([[-1, -1, -1, -1]], dtype = npy.float32)
time1 = time.time()
bottom = int(npy.round(npy.min(tmp[:, 2])))
bif = int(db.getBifurcation(tmp) + 0.5)
up = int(npy.round(npy.max(tmp[:, 2])))
point_vital = [0] * 3
point_vital[0] = tmp[(npy.round(tmp[:, -1]) == 0) & (npy.round(tmp[:, 2]) == bottom)].copy()
point_vital[1] = tmp[(npy.round(tmp[:, -1]) == 1) & (npy.round(tmp[:, 2]) == up)].copy()
point_vital[2] = tmp[(npy.round(tmp[:, -1]) == 2) & (npy.round(tmp[:, 2]) == up)].copy()
autoDetectContour(point_vital[0], 0, bottom, bif, 1, dataset['fix'].getResolution().tolist())
autoDetectContour(point_vital[1], 1, up, bif, -1, dataset['fix'].getResolution().tolist())
autoDetectContour(point_vital[2], 2, up, bif, -1, dataset['fix'].getResolution().tolist())
print ' '
print 'Finish segmentation of MR. '
true_fixed_points = dataset['fix'].pointSet.data['Contour'].copy()
dataset['fix'].pointSet.data['Contour'] = self.new_points
dataset['fix'].pointSet.data['Centerline'] = calCenterlineFromContour(dataset['fix'].pointSet.data)
hybrid = NonrigidHybridRegistration(None)
print 'Register Data %s with Hybrid Method...' % (self.ini.file.name_result[i])
data, point, para = hybrid.register(dataset['fix'], dataset['mov'], regPara = self.regPara, true_fixed_points = true_fixed_points)
print 'Done!'
for k in range(len(self.regPara)):
self.sheet1.write(k + 1, i + 2, float(para[k, 0]))
self.sheet2.write(k + 1, i + 2, float(para[k, 1]))
self.sheet3.write(k + 1, i + 2, float(para[k, 2]))
self.sheet1.write(0, i + 2, self.ini.file.name_result[i])
self.sheet2.write(0, i + 2, self.ini.file.name_result[i])
self.sheet3.write(0, i + 2, self.ini.file.name_result[i])
self.book.save(self.path + self.ini.file.savedir + 'nonrigid' + str(self.first) + '.xls')
del para
del hybrid
del true_fixed_points
def analysis(self, data, point_data_fix = None, area = False):
if point_data_fix is None:
point_data_fix = self.gui.dataModel[data.getFixedIndex()].getPointSet('Contour').copy()
point_data_result = data.getPointSet('Contour').copy()
self.spacing = data.getResolution().tolist()
point_data_fix[:, :2] *= self.spacing[:2]
point_data_result[:, :2] *= self.spacing[:2]
center_data = calCenterlineFromContour({'Contour': point_data_fix})
ind = center_data[:, 2].argsort()
center_data = center_data[ind]
center_data_z = center_data[:, 2].copy()
fixed_bif = db.getBifurcation(center_data)
bif_point = center_data[npy.round(center_data[:, 2]) == fixed_bif - 1]
center_data[:, :3] *= self.spacing[:3]
bif_point[:, :3] *= self.spacing[:3]
spline = [None, None, None]
for cnt in range(3):
spline[cnt] = [None, None, None]
xx = center_data_z[npy.round(center_data[:, -1]) == cnt]
yy = center_data[npy.round(center_data[:, -1]) == cnt]
if cnt > 0:
xx = npy.append(fixed_bif - 1, xx)
yy = npy.append(bif_point, yy, axis = 0)
for i in range(3):
spline[cnt][i] = itp.InterpolatedUnivariateSpline(xx, yy[:, i])
cnt_num = npy.array([0, 0, 0])
mean_dis = npy.array([0.0, 0.0, 0.0])
max_dis = npy.array([0.0, 0.0, 0.0])
square_sum_dis = npy.array([0.0, 0.0, 0.0])
if area:
area_mr = npy.array([0.0, 0.0, 0.0])
area_us = npy.array([0.0, 0.0, 0.0])
for cnt in range(3):
temp_result = point_data_result[npy.where(npy.round(point_data_result[:, -1]) == cnt)]
temp_fix = point_data_fix[npy.where(npy.round(point_data_fix[:, -1]) == cnt)]
if not temp_result.shape[0] or not temp_fix.shape[0]:
continue
zmin = int(npy.max([npy.min(temp_result[:, 2]), npy.min(temp_fix[:, 2])]) + 0.5)
zmax = int(npy.min([npy.max(temp_result[:, 2]), npy.max(temp_fix[:, 2])]) + 0.5)
for z in range(zmin, zmax + 1):
data_fix = temp_fix[npy.where(npy.round(temp_fix[:, 2]) == z)]
data_result = temp_result[npy.where(npy.round(temp_result[:, 2]) == z)]
if data_fix is not None and data_result is not None:
if data_fix.shape[0] == 0 or data_result.shape[0] == 0:
continue
cnt_num[cnt] += 1
#center_fix = npy.mean(data_fix[:, :2], axis = 0)
center_fix, area_fix = calCentroidFromContour(data_fix[:, :2], True)
center_fix = center_fix[0]
#center_result = npy.mean(data_result[:, :2], axis = 0)
center_result, area_result = calCentroidFromContour(data_result[:, :2], True)
center_result = center_result[0]
if area:
area_mr[cnt] += area_fix
area_us[cnt] += area_result
points_fix = getPointsOntheSpline(data_fix, center_fix, 900)
points_result = getPointsOntheSpline(data_result, center_result, 900)
normal = npy.array([None, None, None])
for i in range(3):
normal[i] = spline[cnt][i].derivatives(z)[1]
w1 = normal[2] ** 2 / npy.sum(normal ** 2) # cos(alpha) ^ 2
theta0 = npy.arctan2(normal[1], normal[0])
i = j = 0
for k in range(-44, 46):
angle = k * 4 / 180.0 * npy.pi
while i < 900 and points_fix[i, 2] < angle:
i += 1
if i == 900 or (i > 0 and angle - points_fix[i - 1, 2] < points_fix[i, 2] - angle):
ind_fix = i - 1
else:
ind_fix = i
while j < 900 and points_result[j, 2] < angle:
j += 1
if j == 900 or (j > 0 and angle - points_result[j - 1, 2] < points_result[j, 2] - angle):
ind_result = j - 1
else:
ind_result = j
weigh = npy.sqrt(npy.sin(angle - theta0) ** 2 + npy.cos(angle - theta0) ** 2 / w1)
temp_dis = npy.hypot(points_fix[ind_fix, 0] - points_result[ind_result, 0], points_fix[ind_fix, 1] - points_result[ind_result, 1]) / weigh
#if area:
# temp_dis /= max([area_result / area_fix, area_fix / area_result])
max_dis[cnt] = npy.max([max_dis[cnt], temp_dis])
mean_dis[cnt] += temp_dis
cnt_total = npy.sum(cnt_num)
mean_dis /= 90
if area:
for cnt in range(3):
if area_mr[cnt] < area_us[cnt]:
rate = area_us[cnt] / area_mr[cnt]
else:
rate = area_us[cnt] / area_mr[cnt]
mean_dis[cnt] /= rate
mean_whole = npy.sum(mean_dis)
mean_dis /= cnt_num
mean_dis[mean_dis != mean_dis] = 0 # Replace the NAN in the mean distance
if self.gui is not None:
message = "Error on Vessel 0: %0.2fmm (Total %d slices)\nError on Vessel 1: %0.2fmm (Total %d slices)\nError on Vessel 2: %0.2fmm (Total %d slices)\nWhole Error: %0.2fmm (Total %d slices)\n" \
% (mean_dis[0], cnt_num[0], mean_dis[1], cnt_num[1], mean_dis[2], cnt_num[2], mean_whole / cnt_total, cnt_total) + \
"-----------------------------------------------------------------------------\n" + \
"Max Error on Vessel 0: %0.2fmm\nMax Error on Vessel 1: %0.2fmm\nMax Error on Vessel 2: %0.2fmm\nTotal Max Error: %0.2fmm" \
% (max_dis[0], max_dis[1], max_dis[2], npy.max(max_dis));
self.gui.showErrorMessage("Weighted Contour Registration Error", message)
return mean_dis, mean_whole / cnt_total, max_dis, npy.max(max_dis)
def process(self, dataset, i):
# ICP with centerline
print 'Register Data %s with ICP...' % self.ini.file.name_result[i]
tmp = dataset['mov'].pointSet.data['Contour'].copy()
for sd in range(0, 16):
mean_dis_all = npy.zeros([4, 3], dtype=npy.float32)
mean_whole_all = npy.zeros([4, 1], dtype=npy.float32)
if sd > 0:
repeat = self.repeat
else:
repeat = 1
for i in range(repeat):
dataset['mov'].pointSet.data['Contour'] = AddNoise(
tmp,
float(sd) / 5)
dataset['mov'].pointSet.data[
'Centerline'] = calCenterlineFromContour(
dataset['mov'].pointSet.data)
# Centerline label
data, point, para = self.icp.register(dataset['fix'],
dataset['mov'], 1)
resultData = db.ResultData(
data, db.ImageInfo(dataset['fix'].info.data), point)
resultData.info.addData('fix', 1)
resultData.info.addData('move', 2)
resultData.info.addData('transform', para)
mean_dis, mean_whole, max_dis, max_whole = self.surfaceerror.analysis(
resultData, dataset['fix'].getPointSet('Contour').copy(),
dataset['mov'].getPointSet('Contour').copy(),
dataset['mov'].getPointSet('Mask').copy(),
dataset['mov'].getResolution().tolist())
mean_dis_all[0, :] += mean_dis
mean_whole_all[0] += mean_whole
del data, point, resultData, para
# Contour label
data, point, para = self.icp.register(dataset['fix'],
dataset['mov'],
0,
op=True)
resultData = db.ResultData(
data, db.ImageInfo(dataset['fix'].info.data), point)
resultData.info.addData('fix', 1)
resultData.info.addData('move', 2)
resultData.info.addData('transform', para)
mean_dis, mean_whole, max_dis, max_whole = self.surfaceerror.analysis(
resultData, dataset['fix'].getPointSet('Contour').copy(),
dataset['mov'].getPointSet('Contour').copy(),
dataset['mov'].getPointSet('Mask').copy(),
dataset['mov'].getResolution().tolist())
mean_dis_all[1, :] += mean_dis
mean_whole_all[1] += mean_whole
del data, point, resultData, para
# Centerline no-label
data, point, para = self.icp.register(dataset['fix'],
dataset['mov'],
1,
op=True)
resultData = db.ResultData(
data, db.ImageInfo(dataset['fix'].info.data), point)
resultData.info.addData('fix', 1)
resultData.info.addData('move', 2)
resultData.info.addData('transform', para)
mean_dis, mean_whole, max_dis, max_whole = self.surfaceerror.analysis(
resultData, dataset['fix'].getPointSet('Contour').copy(),
dataset['mov'].getPointSet('Contour').copy(),
dataset['mov'].getPointSet('Mask').copy(),
dataset['mov'].getResolution().tolist())
mean_dis_all[2, :] += mean_dis
mean_whole_all[2] += mean_whole
del data, point, resultData, para
# Contour no-label
data, point, para = self.icp.register(dataset['fix'],
dataset['mov'], 0)
resultData = db.ResultData(
data, db.ImageInfo(dataset['fix'].info.data), point)
resultData.info.addData('fix', 1)
resultData.info.addData('move', 2)
resultData.info.addData('transform', para)
mean_dis, mean_whole, max_dis, max_whole = self.surfaceerror.analysis(
resultData, dataset['fix'].getPointSet('Contour').copy(),
dataset['mov'].getPointSet('Contour').copy(),
dataset['mov'].getPointSet('Mask').copy(),
dataset['mov'].getResolution().tolist())
mean_dis_all[3, :] += mean_dis
mean_whole_all[3] += mean_whole
del data, point, resultData, para
sys.stdout.write(str(i) + ',')
sys.stdout.flush()
mean_dis_all /= repeat
mean_whole_all /= repeat
print ' '
print 'Noise level %fmm Done!' % (float(sd) / 5)
print 'Contour Error Done! Whole mean is %0.2fmm vs %0.2fmm.' % (
mean_whole_all[0], mean_whole_all[1])
for i in range(4):
self.error[i, sd, :3] += mean_dis_all[i, :]
self.error[i, sd, 3] += mean_whole_all[i]
def process(self, dataset, i):
def autoDetectContour(point, cnt, start, end, delta, res, type):
self.new_points = npy.append(self.new_points, point, 0)
points = point[:, :-2]
d = 20
count = 0
for i in range(start + delta, end + delta, delta):
center = calCentroidFromContour(points).reshape(2)
image = dataset[type].getData()[i, :, :].transpose().copy()
image = (image - npy.min(image)) / (npy.max(image) - npy.min(image)) * 255
down = npy.max([npy.ceil(center[0] - d / res[0]), 0])
up = npy.min([npy.floor(center[0] + d / res[0]), image.shape[0]])
left = npy.max([npy.ceil(center[1] - d / res[1]), 0])
right = npy.min([npy.floor(center[1] + d / res[1]), image.shape[1]])
crop_image = image[down : up, left : right]
center -= [down, left]
result = ac_segmentation(center, crop_image)
a1 = ac_area(points.transpose(), image.shape)
a2 = ac_area(result, crop_image.shape)
rate = a2 * 1.0 / a1
if rate >= min(1.5 + count * 0.2, 2.1) or rate <= 0.7:
temp_array = points.copy()
if cnt != 1 and rate > 0.7:
count += 1
else:
temp_array = result.transpose().copy()
temp_array[:, :2] += [down, left]
count = 0
points = temp_array.copy()
temp_array = npy.insert(temp_array, 2, [[i], [cnt]], 1)
self.new_points = npy.append(self.new_points, temp_array, 0)
sys.stdout.write(str(i) + ',')
sys.stdout.flush()
print ' '
# Segmentation of data
print 'Segment Data %s...' % self.ini.file.name_result[i]
tmp = dataset['fix'].pointSet.data['Contour']
tmp = tmp[tmp[:, 0] >= 0]
self.new_points = npy.array([[-1, -1, -1, -1]], dtype = npy.float32)
bottom = int(npy.round(npy.min(tmp[:, 2])))
bif = int(db.getBifurcation(tmp) + 0.5)
up = int(npy.round(npy.max(tmp[:, 2])))
bottom += (bif - bottom) / 2
up -= (up - bif) / 2
point_vital = [0] * 3
point_vital[0] = tmp[(npy.round(tmp[:, -1]) == 0) & (npy.round(tmp[:, 2]) == bottom)].copy()
point_vital[1] = tmp[(npy.round(tmp[:, -1]) == 1) & (npy.round(tmp[:, 2]) == up)].copy()
point_vital[2] = tmp[(npy.round(tmp[:, -1]) == 2) & (npy.round(tmp[:, 2]) == up)].copy()
autoDetectContour(point_vital[0], 0, bottom, bif, 1, dataset['fix'].getResolution().tolist(), 'fix')
autoDetectContour(point_vital[1], 1, up, bif, -1, dataset['fix'].getResolution().tolist(), 'fix')
autoDetectContour(point_vital[2], 2, up, bif, -1, dataset['fix'].getResolution().tolist(), 'fix')
print ' '
print 'Finish segmentation for fix data. '
pointset = {'Contour': self.new_points}
dataset['fix'].pointSet.data['Centerline'] = calCenterlineFromContour(pointset)
self.new_points_fix = self.new_points.copy()
# For mov data
tmp = dataset['mov'].pointSet.data['Contour']
tmp = tmp[tmp[:, 0] >= 0]
self.new_points = npy.array([[-1, -1, -1, -1]], dtype = npy.float32)
bottom = int(npy.round(npy.min(tmp[:, 2])))
bif = int(db.getBifurcation(tmp) + 0.5)
up = int(npy.round(npy.max(tmp[:, 2])))
bottom += (bif - bottom) / 2
up -= (up - bif) / 2
point_vital = [0] * 3
point_vital[0] = tmp[(npy.round(tmp[:, -1]) == 0) & (npy.round(tmp[:, 2]) == bottom)].copy()
point_vital[1] = tmp[(npy.round(tmp[:, -1]) == 1) & (npy.round(tmp[:, 2]) == up)].copy()
point_vital[2] = tmp[(npy.round(tmp[:, -1]) == 2) & (npy.round(tmp[:, 2]) == up)].copy()
autoDetectContour(point_vital[0], 0, bottom, bif, 1, dataset['mov'].getResolution().tolist(), 'mov')
autoDetectContour(point_vital[1], 1, up, bif, -1, dataset['mov'].getResolution().tolist(), 'mov')
autoDetectContour(point_vital[2], 2, up, bif, -1, dataset['mov'].getResolution().tolist(), 'mov')
print ' '
print 'Finish segmentation for mov data. '
pointset = {'Contour': self.new_points}
dataset['mov'].pointSet.data['Centerline'] = calCenterlineFromContour(pointset)
self.new_points_mov = self.new_points.copy()
# ICP with centerline without label
print 'Register Data %s with ICP(centerline) without label...' % self.ini.file.name_result[i]
data, point, para = self.icp.register(dataset['fix'], dataset['mov'], 1, op = True)
resultData = db.ResultData(data, db.ImageInfo(dataset['fix'].info.data), point)
resultData.info.addData('fix', 1)
resultData.info.addData('move', 2)
resultData.info.addData('transform', para)
print 'Done!'
mean_dis, mean_whole, max_dis, max_whole = self.surfaceerror.analysis(resultData, dataset['fix'].getPointSet('Contour').copy(), dataset['mov'].getPointSet('Contour').copy(), dataset['mov'].getPointSet('Mask').copy(), dataset['mov'].getResolution().tolist())
print 'Contour Error Done! Whole mean is %0.2fmm.' % mean_whole
dice_index, dice_index_all = self.areaerror.analysis(resultData, dataset['fix'].getPointSet('Contour').copy())
print 'Area Error Done! Whole Dice index is %0.3f.' % dice_index_all
self.sheet2.write(0, i + 2, self.ini.file.name_result[i])
for j in range(3):
self.sheet2.write(j + 1, i + 2, mean_dis[j])
self.sheet2.write(j + 5, i + 2, max_dis[j])
self.sheet2.write(j + 9, i + 2, dice_index[j])
self.sheet2.write(4, i + 2, mean_whole)
self.sheet2.write(8, i + 2, max_whole)
self.sheet2.write(12, i + 2, dice_index_all)
self.book.save(self.path + self.ini.file.savedir + 'multicontrast_seg_feature.xls')
del data, point, resultData
# ICP with centerline with label
print 'Register Data %s with ICP(centerline) with label...' % self.ini.file.name_result[i]
data, point, para = self.icp.register(dataset['fix'], dataset['mov'], 1, op = False)
resultData = db.ResultData(data, db.ImageInfo(dataset['fix'].info.data), point)
resultData.info.addData('fix', 1)
resultData.info.addData('move', 2)
resultData.info.addData('transform', para)
print 'Done!'
mean_dis, mean_whole, max_dis, max_whole = self.surfaceerror.analysis(resultData, dataset['fix'].getPointSet('Contour').copy(), dataset['mov'].getPointSet('Contour').copy(), dataset['mov'].getPointSet('Mask').copy(), dataset['mov'].getResolution().tolist())
print 'Contour Error Done! Whole mean is %0.2fmm.' % mean_whole
dice_index, dice_index_all = self.areaerror.analysis(resultData, dataset['fix'].getPointSet('Contour').copy())
print 'Area Error Done! Whole Dice index is %0.3f.' % dice_index_all
self.sheet4.write(0, i + 2, self.ini.file.name_result[i])
for j in range(3):
self.sheet4.write(j + 1, i + 2, mean_dis[j])
self.sheet4.write(j + 5, i + 2, max_dis[j])
self.sheet4.write(j + 9, i + 2, dice_index[j])
self.sheet4.write(4, i + 2, mean_whole)
self.sheet4.write(8, i + 2, max_whole)
self.sheet4.write(12, i + 2, dice_index_all)
self.book.save(self.path + self.ini.file.savedir + 'multicontrast_seg_feature.xls')
del data, point, resultData
fix_points = dataset['fix'].getPointSet('Contour').copy()
dataset['fix'].pointSet.data['Contour'] = self.new_points_fix
mov_points = dataset['mov'].getPointSet('Contour').copy()
dataset['mov'].pointSet.data['Contour'] = self.new_points_mov
print 'Saving Data %s...' % self.ini.file.name_result[i]
db.saveMatData(self.path + self.ini.file.savedir + self.ini.file.name_result[i] + '_snap.mat', [dataset['fix']], 0)
db.saveMatData(self.path + self.ini.file.savedir + self.ini.file.name_result[i] + '_merge.mat', [dataset['mov']], 0)
print 'Done!'
# ICP with contour without label
print 'Register Data %s with ICP(contour) without label...' % self.ini.file.name_result[i]
data, point, para = self.icp.register(dataset['fix'], dataset['mov'], 0, op = False)
resultData = db.ResultData(data, db.ImageInfo(dataset['fix'].info.data), point)
resultData.info.addData('fix', 1)
resultData.info.addData('move', 2)
resultData.info.addData('transform', para)
print 'Done!'
mean_dis, mean_whole, max_dis, max_whole = self.surfaceerror.analysis(resultData, fix_points.copy(), mov_points.copy(), dataset['mov'].getPointSet('Mask').copy(), dataset['mov'].getResolution().tolist())
print 'Contour Error Done! Whole mean is %0.2fmm.' % mean_whole
para = resultData.info.getData('transform')
R = ml.mat(para[:9]).reshape(3, 3)
T = ml.mat(para[9:12]).T
T = R.I * T
T = -T
tmp_con, result_center_points = util.resliceTheResultPoints(mov_points, None, 20, dataset['mov'].getResolution().tolist(),
dataset['fix'].getResolution().tolist(), False, R, T)
resultData.pointSet.data['Contour'] = tmp_con
dice_index, dice_index_all = self.areaerror.analysis(resultData, fix_points.copy())
print 'Area Error Done! Whole Dice index is %0.3f.' % dice_index_all
self.sheet1.write(0, i + 2, self.ini.file.name_result[i])
for j in range(3):
self.sheet1.write(j + 1, i + 2, mean_dis[j])
self.sheet1.write(j + 5, i + 2, max_dis[j])
self.sheet1.write(j + 9, i + 2, dice_index[j])
self.sheet1.write(4, i + 2, mean_whole)
self.sheet1.write(8, i + 2, max_whole)
self.sheet1.write(12, i + 2, dice_index_all)
self.book.save(self.path + self.ini.file.savedir + 'multicontrast_seg_feature.xls')
del data, point, resultData
# ICP with contour with label
print 'Register Data %s with ICP(contour) with label...' % self.ini.file.name_result[i]
data, point, para = self.icp.register(dataset['fix'], dataset['mov'], 0, op = True)
resultData = db.ResultData(data, db.ImageInfo(dataset['fix'].info.data), point)
resultData.info.addData('fix', 1)
resultData.info.addData('move', 2)
resultData.info.addData('transform', para)
print 'Done!'
mean_dis, mean_whole, max_dis, max_whole = self.surfaceerror.analysis(resultData, fix_points.copy(), mov_points.copy(), dataset['mov'].getPointSet('Mask').copy(), dataset['mov'].getResolution().tolist())
print 'Contour Error Done! Whole mean is %0.2fmm.' % mean_whole
para = resultData.info.getData('transform')
R = ml.mat(para[:9]).reshape(3, 3)
T = ml.mat(para[9:12]).T
T = R.I * T
T = -T
tmp_con, result_center_points = util.resliceTheResultPoints(mov_points, None, 20, dataset['mov'].getResolution().tolist(),
dataset['fix'].getResolution().tolist(), False, R, T)
resultData.pointSet.data['Contour'] = tmp_con
dice_index, dice_index_all = self.areaerror.analysis(resultData, fix_points.copy())
print 'Area Error Done! Whole Dice index is %0.3f.' % dice_index_all
self.sheet3.write(0, i + 2, self.ini.file.name_result[i])
for j in range(3):
self.sheet3.write(j + 1, i + 2, mean_dis[j])
self.sheet3.write(j + 5, i + 2, max_dis[j])
self.sheet3.write(j + 9, i + 2, dice_index[j])
self.sheet3.write(4, i + 2, mean_whole)
self.sheet3.write(8, i + 2, max_whole)
self.sheet3.write(12, i + 2, dice_index_all)
self.book.save(self.path + self.ini.file.savedir + 'multicontrast_seg_feature.xls')
del data, point, resultData
del self.new_points, fix_points, self.new_points_fix, self.new_points_mov, mov_points, tmp_con, result_center_points
def process(self, dataset, i):
def autoDetectContour(point, cnt, start, end, delta):
self.new_points = npy.append(self.new_points, point, 0)
points = point[:, :-2]
count = 0
for j in range(start + delta, end + delta, delta):
center = calCentroidFromContour(points).reshape(2)
image = dataset['fix'].getData()[j, :, :].transpose().copy()
image = (image - npy.min(image)) / (npy.max(image) -
npy.min(image)) * 255
result = ac_segmentation(center, image)
a1 = ac_area(points.transpose(), image.shape)
a2 = ac_area(result, image.shape)
rate = a2 * 1.0 / a1
if rate >= min(1.5 + count * 0.2, 2.1) or rate <= 0.7:
temp_array = points.copy()
if cnt != 1 and rate > 0.7:
count += 1
else:
temp_array = result.transpose().copy()
count = 0
points = temp_array.copy()
temp_array = npy.insert(temp_array, 2, [[j], [cnt]], 1)
self.new_points = npy.append(self.new_points, temp_array, 0)
sys.stdout.write(str(j) + ',')
sys.stdout.flush()
print ' '
print 'Segment Data %s...' % self.ini.file.name_result[i]
tmp = dataset['fix'].pointSet.data['Contour']
tmp = tmp[tmp[:, 0] >= 0]
self.new_points = npy.array([[-1, -1, -1, -1]], dtype=npy.float32)
time1 = time.time()
bottom = int(npy.round(npy.min(tmp[:, 2])))
bif = int(db.getBifurcation(tmp) + 0.5)
up = int(npy.round(npy.max(tmp[:, 2])))
point_vital = [0] * 3
point_vital[0] = tmp[(npy.round(tmp[:, -1]) == 0)
& (npy.round(tmp[:, 2]) == bottom)].copy()
point_vital[1] = tmp[(npy.round(tmp[:, -1]) == 1)
& (npy.round(tmp[:, 2]) == up)].copy()
point_vital[2] = tmp[(npy.round(tmp[:, -1]) == 2)
& (npy.round(tmp[:, 2]) == up)].copy()
autoDetectContour(point_vital[0], 0, bottom, bif, 1)
autoDetectContour(point_vital[1], 1, up, bif, -1)
autoDetectContour(point_vital[2], 2, up, bif, -1)
time2 = time.time()
'''
# Use centerline for contour
j = 0
for center in tmp:
image = dataset['fix'].getData()[npy.round(center[2]), :, :].transpose().copy()
image = (image - npy.min(image)) / (npy.max(image) - npy.min(image)) * 255
result = ac_segmentation(center[:2], image)
point_array = npy.insert(result.transpose(), 2, [[center[2]],[center[3]]], axis = 1)
new_points = npy.append(new_points, point_array, 0)
j += 1
if j % 10 == 0:
sys.stdout.write(str(j) + ',')
sys.stdout.flush()
'''
print ' '
print 'Done! Time for segmentation is %0.2fs' % (time2 - time1)
pointset = {'Contour': self.new_points}
pointset['Centerline'] = calCenterlineFromContour(pointset)
print 'Saving Data %s...' % self.ini.file.name_result[i]
new_data = db.BasicData(dataset['fix'].data,
db.ImageInfo(dataset['fix'].info.data),
pointset)
db.saveMatData(
self.savepath + self.ini.file.name_result[i] + '_mr.mat',
[new_data], 0)
print 'Done!'
mean_dis, mean_whole, max_dis, max_whole = self.contourerror.analysis(
new_data, dataset['fix'].getPointSet('Contour').copy())
print 'Contour Error Done! Whole mean is %0.2fmm.' % mean_whole
self.sheet1.write(0, i + 2, self.ini.file.name_result[i])
for j in range(3):
self.sheet1.write(j + 1, i + 2, mean_dis[j])
self.sheet1.write(4, i + 2, mean_whole)
self.sheet1.write(5, i + 2, time2 - time1)
self.book.save(self.path + self.ini.file.savedir +
'Test_segmentation_final_refined.xls')
# ICP with centerline
print 'Register Data %s with ICP(centerline)...' % self.ini.file.name_result[
i]
time1 = time.time()
data, point, para = self.icp.register(new_data, dataset['mov'], 1)
time2 = time.time()
print 'Done! Time for registration is %0.2fs' % (time2 - time1)
resultData = db.ResultData(data,
db.ImageInfo(dataset['fix'].info.data),
point)
resultData.info.addData('fix', 1)
resultData.info.addData('move', 2)
resultData.info.addData('transform', para)
mean_dis, mean_whole, max_dis, max_whole = self.contourerror.analysis(
resultData, dataset['fix'].getPointSet('Contour').copy())
print 'Contour Error Done! Whole mean is %0.2fmm.' % mean_whole
dice_index, dice_index_all = self.areaerror.analysis(
resultData, dataset['fix'].getPointSet('Contour').copy())
print 'Area Error Done! Whole Dice index is %0.3f.' % dice_index_all
self.sheet2.write(0, i + 2, self.ini.file.name_result[i])
for j in range(3):
self.sheet2.write(j + 1, i + 2, mean_dis[j])
self.sheet2.write(j + 5, i + 2, max_dis[j])
self.sheet2.write(j + 9, i + 2, dice_index[j])
self.sheet2.write(4, i + 2, mean_whole)
self.sheet2.write(8, i + 2, max_whole)
self.sheet2.write(12, i + 2, dice_index_all)
self.sheet2.write(13, i + 2, time2 - time1)
self.book.save(self.path + self.ini.file.savedir +
'Test_segmentation_final_refined2.xls')
del data, point, resultData