def create_sliceinfo_w(images_fn,
labels_fn,
regint_fn,
kernels,
i,
slice_no=None):
# figure out the biggest slice
if slice_no == None:
slice_no = find_biggest_slice(path + labels_fn[i])
print("Creating Weighted Slice Info... {}/{}: Slice {}".format(
i + 1, len(images_fn), slice_no))
# get the slice, label, and associated orientations
slice_im, slice_im_or = get_nifti_slice(path + images_fn[i], slice_no)
slice_lb, slice_lb_or = get_nifti_slice(path + labels_fn[i], slice_no)
slice_ro, slice_ro_or = get_nifti_slice(path + regint_fn[i], slice_no)
# if crop, we crop the image down
if crop:
crop_x, crop_y = crop
start_x = slice_im.shape[0] / 2 - crop_x / 2
end_x = start_x + crop_x
start_y = slice_im.shape[1] / 2 - crop_y / 2
end_y = start_y + crop_y
slice_im = slice_im[start_x:end_x, start_y:end_y]
slice_lb = slice_lb[start_x:end_x, start_y:end_y]
slice_ro = slice_ro[start_x:end_x, start_y:end_y]
# figure out the principal patches
pc_payload = (slice_im, slice_lb)
patches_m_pc, patches_n_pc, vals_m, vals_n = create_pc_patches_w(
*pc_payload)
# compute gabor features for the patches
feats_m = []
feats_n = []
intens_m = []
intens_n = []
for patch in patches_m_pc:
feats_m.append(compute_feats(patch, kernels))
intens_m.append(compute_intens(patch))
for patch in patches_n_pc:
feats_n.append(compute_feats(patch, kernels))
intens_n.append(compute_intens(patch))
# package it into a SliceInfo object
si_payload = (images_fn[i], slice_no, slice_im, slice_im_or, slice_lb,
slice_lb_or, slice_ro, slice_ro_or, patches_m_pc,
patches_n_pc, feats_m, feats_n, intens_m, intens_n, vals_m,
vals_n)
return SliceInfo(*si_payload)
def create_sliceinfo_w(images_fn, labels_fn, regint_fn, kernels, i, slice_no=None):
# figure out the biggest slice
if slice_no == None:
slice_no = find_biggest_slice(path + labels_fn[i])
print("Creating Weighted Slice Info... {}/{}: Slice {}".format(i+1, len(images_fn), slice_no))
# get the slice, label, and associated orientations
slice_im, slice_im_or = get_nifti_slice(path + images_fn[i], slice_no)
slice_lb, slice_lb_or = get_nifti_slice(path + labels_fn[i], slice_no)
slice_ro, slice_ro_or = get_nifti_slice(path + regint_fn[i], slice_no)
# if crop, we crop the image down
if crop:
crop_x, crop_y = crop
start_x = slice_im.shape[0] / 2 - crop_x / 2
end_x = start_x + crop_x
start_y = slice_im.shape[1] / 2 - crop_y / 2
end_y = start_y + crop_y
slice_im = slice_im[start_x:end_x, start_y:end_y]
slice_lb = slice_lb[start_x:end_x, start_y:end_y]
slice_ro = slice_ro[start_x:end_x, start_y:end_y]
# figure out the principal patches
pc_payload = (slice_im, slice_lb)
patches_m_pc, patches_n_pc, vals_m, vals_n = create_pc_patches_w(*pc_payload)
# compute gabor features for the patches
feats_m = []
feats_n = []
intens_m = []
intens_n = []
for patch in patches_m_pc:
feats_m.append(compute_feats(patch, kernels))
intens_m.append(compute_intens(patch))
for patch in patches_n_pc:
feats_n.append(compute_feats(patch, kernels))
intens_n.append(compute_intens(patch))
# package it into a SliceInfo object
si_payload = (images_fn[i], slice_no,
slice_im, slice_im_or,
slice_lb, slice_lb_or,
slice_ro, slice_ro_or,
patches_m_pc, patches_n_pc,
feats_m, feats_n,
intens_m, intens_n,
vals_m, vals_n)
return SliceInfo(*si_payload)
def classify_patch_group(fn_rf, fn_kern, fn_patch_info):
t0 = time()
RF = dill.load(open(fn_rf, 'rb')) # unpack the RF classifier
kernels = dill.load(open(fn_kern, 'rb')) # unpack the kernels
patch_info = dill.load(open(fn_patch_info, 'rb')) # unpack the patch info
a, b, c = patch_info[0] # get the bounds of the set
patches_a = patch_info[1] # grab the set to classify
patches_r = patch_info[2] # grab the set to compare with (atlas mask)
results = []
for i, patch in enumerate(patches_a): # go through each patch
if np.all(patches_r[i]): # if the patch is entirely masked
feat = compute_feats(patch, kernels).flatten().reshape(1, -1)
intens = np.array(compute_intens(patch)).flatten().reshape(1, -1)
feat = np.concatenate((feat, intens), axis=1)
prediction = RF.predict(feat)
#print("Classifying patch {}/{}: {}".format(i, len(patches), prediction))
results.append(np.full(patch.shape, prediction))
else: # the associated ROI patch is totally zero
results.append(np.zeros(patch.shape))
dt = time() - t0
print("Classified group {}-{}/{} in {:.2f} time".format(a, b, c, dt))
return results
def classify_patch_group(fn_rf, fn_kern, fn_patch_info):
t0 = time()
RF = dill.load(open(fn_rf, "rb")) # unpack the RF classifier
kernels = dill.load(open(fn_kern, "rb")) # unpack the kernels
patch_info = dill.load(open(fn_patch_info, "rb")) # unpack the patch info
a, b, c = patch_info[0] # get the bounds of the set
patches_a = patch_info[1] # grab the set to classify
patches_r = patch_info[2] # grab the set to compare with (atlas mask)
results = []
for i, patch in enumerate(patches_a): # go through each patch
if np.all(patches_r[i]): # if the patch is entirely masked
feat = compute_feats(patch, kernels).flatten().reshape(1, -1)
intens = np.array(compute_intens(patch)).flatten().reshape(1, -1)
feat = np.concatenate((feat, intens), axis=1)
prediction = RF.predict(feat)
# print("Classifying patch {}/{}: {}".format(i, len(patches), prediction))
results.append(np.full(patch.shape, prediction))
else: # the associated ROI patch is totally zero
results.append(np.zeros(patch.shape))
dt = time() - t0
print("Classified group {}-{}/{} in {:.2f} time".format(a, b, c, dt))
return results