def visualize_prediction( algorithms, pred_results, M, labels, output_prefix=""):
visualize_prediction.counter += 1
for m in M:
metadata = m
for algorithm in algorithms:
alg_name = algorithm[2]
mean_preds = np.sum(np.array(pred_results[alg_name])-np.array(labels), axis=0)
sd_preds = np.std(np.array(pred_results[alg_name])-np.array(labels), axis=0)
scalars = np.abs(mean_preds * 10. / np.max(mean_preds))
petacc_root = "/datagrid/Medical/microscopy/petacc3"
image_name = os.path.basename(metadata[0])[:8]
patch_size = int(os.path.basename(metadata[0]).split('_')[3][1:])
image_path = petacc_root + "/batch_2/" + image_name + ".ndpi"
if not os.path.exists(image_path):
image_path = petacc_root + "/batch_1/" + image_name + ".ndpi"
si = NDPISlideImage(image_path, None)
patchvis = si.get_patch_visualization(5, metadata[2], patch_size,
scalars=scalars[m[3]:m[3]+len(m[2])],
line_thickness=2,
show=False )
cv2.imwrite("/tmp/{}_l0_vis_".format(output_prefix)+image_name+"_"+alg_name+"features_{:d}.png".format(visualize_prediction.counter), patchvis)
def visualize_prediction( feature_name, pred_results, M, labels, train_len=0, output_prefix=""):
data_counter = 0
for m in M:
metadata = m
# mean_preds = np.sum(np.array(pred_results[alg_name])-np.array(labels), axis=0)
scalars = np.abs(pred_results * 100. / np.max(pred_results))
petacc_root = "/datagrid/Medical/microscopy/petacc3"
image_name = os.path.basename(metadata[0])[:8]
patch_size = 1024
image_path = petacc_root + "/batch_2/" + image_name + ".ndpi"
if not os.path.exists(image_path):
image_path = petacc_root + "/batch_1/" + image_name + ".ndpi"
si = NDPISlideImage(image_path, None)
patchvis = si.get_patch_visualization(6, metadata[2], patch_size,
scalars=scalars[m[3]:m[3]+len(m[2])],
line_thickness=1,
show=False, filled=True)
# if data_counter < train_len:
# cv2.putText(patchvis, 'test', (40, 40), cv2.FONT_HERSHEY_PLAIN, 1, (255, 0, 0), 2, cv2.LINE_AA)
# else:
# cv2.putText(patchvis, 'vali', (40, 40), cv2.FONT_HERSHEY_PLAIN, 1, (0, 0, 255), 2, cv2.LINE_AA)
cv2.imwrite("{}_{}_{}".format(output_prefix, image_name, feature_name) + ".png", patchvis)
data_counter += 1
def visualize_prediction( algorithms, pred_results, M, labels, train_len=0, output_prefix=""):
visualize_prediction.counter += 1
data_counter = 0
validation_offset = 0
imtype = ''
for m in M:
metadata = m
for algorithm in algorithms:
alg_name = algorithm[2]
# mean_preds = np.sum(np.array(pred_results[alg_name])-np.array(labels), axis=0)
mean_preds = np.sum(np.array(pred_results[alg_name]), axis=0)
max_preds = np.max(np.array(pred_results[alg_name]), axis=0)
min_preds = np.min(np.array(pred_results[alg_name]), axis=0)
sd_preds = np.sum(np.array(pred_results[alg_name]), axis=0)
# sd_preds = np.std(np.array(pred_results[alg_name])-np.array(labels), axis=0)
scalars = np.abs(mean_preds * 100. / np.max(mean_preds))
border_scalars = np.zeros_like(scalars)
for li, label in enumerate(labels):
if label > 0:
border_scalars[li] = 100 * min_preds[li]
else:
border_scalars[li] = 100 * max_preds[li]
petacc_root = "/datagrid/Medical/microscopy/petacc3"
image_name = os.path.basename(metadata[0])[:8]
patch_size = int(os.path.basename(metadata[0]).split('_')[3][1:])
image_path = petacc_root + "/batch_2/" + image_name + ".ndpi"
if not os.path.exists(image_path):
image_path = petacc_root + "/batch_1/" + image_name + ".ndpi"
si = NDPISlideImage(image_path, None)
patchvis = si.get_patch_visualization(6, metadata[2], patch_size,
scalars=scalars[validation_offset+m[3]:validation_offset+m[3]+len(m[2])],
border_scalars=border_scalars[validation_offset+m[3]:validation_offset+m[3]+len(m[2])],
line_thickness=1,
show=False, filled=True)
if data_counter < train_len:
cv2.putText(patchvis, 'test', (40, 40), cv2.FONT_HERSHEY_PLAIN, 1, (255, 0, 0), 2, cv2.LINE_AA)
imtype='test'
else:
cv2.putText(patchvis, 'vali', (40, 40), cv2.FONT_HERSHEY_PLAIN, 1, (0, 0, 255), 2, cv2.LINE_AA)
imtype='vali'
data_counter += 1
if data_counter == train_len:
validation_offset = m[3]+len(m[2])
cv2.imwrite("/local/temporary/clf_vis/{}".format(output_prefix)+image_name+"_{}_".format(imtype)+alg_name+"features_{:d}.png".format(visualize_prediction.counter), patchvis)
if line_count == 0:
line_count += 1
elif line_count < 84:
split_row = row[0].split(';')
if split_row[1] == 'None' and cmdline_args.skip:
line_count += 1
continue
image_path = (cmdline_args.folder + "\\" + split_row[0] + ".ndpi").replace("/", "\\").replace("\\\\",
"\\")
annotation_path = (cmdline_args.folder + "\\" + split_row[1]).replace("/", "\\").replace("\\\\", "\\")
mask_path = (cmdline_args.folder + "\\" + split_row[2]).replace("/", "\\").replace("\\\\", "\\")
cat_muc = split_row[3]
cat_ser = split_row[4]
cat_cro = split_row[5]
si = NDPISlideImage(image_path=image_path,
tissue_mask_path=mask_path, tumor_annotation_file=annotation_path)
p = si.get_annotated_patches(extract_level=cmdline_args.extract_level,
min_coverage_extraction=cmdline_args.min_coverage,
min_tumor_coverage=0.9, p_size=cmdline_args.patch_size, p_shift=p_shift)
name = split_row[0].split("/")[1].split("-")[0]
patch_count = 0
all_count = 1
tucount = 0
nocount = 0
bocount = 0
for patch in p:
patch_array = si.load_patch(patch[0], (cmdline_args.patch_size, cmdline_args.patch_size))
image = Image.fromarray(patch_array)
if augmentation:
images = [image, image.transpose(Image.FLIP_LEFT_RIGHT), image.transpose(Image.FLIP_TOP_BOTTOM),
[0.4895436, 0.74380669, 0.50788103],
[0.53518641, 0.76756465, 0.35352657]])
modulation_matrix = None
if cmdline_args.convolution_matrix is not None:
rgb_from_her = eval(cmdline_args.convolution_matrix)
print('Retrieved color-mixing matrix: \n ' + str(rgb_from_her))
her_from_rgb = np.linalg.inv(rgb_from_her)
if cmdline_args.modulation_matrix is not None:
modulation_matrix = eval(cmdline_args.modulation_matrix)
print('Retrieved modulation matrix: \n ' + str(modulation_matrix))
if '.ndpi' in cmdline_args.image:
si = NDPISlideImage(image_path=cmdline_args.image,
tissue_mask_path=cmdline_args.mask)
is_camelyon = False
rgb_from_her = np.array([[0.47680668, 0.54035088, 0.33853616],
[0.33593786, 0.74302103, 0.58208704],
[0.42269833, 0.80876244, 0.37791299]])
p_shift = cmdline_args.patch_size
p = si.get_annotated_patches(
extract_level=cmdline_args.extract_level,
min_coverage_extraction=cmdline_args.min_coverage,
min_tumor_coverage=0.6,
p_size=cmdline_args.patch_size,
p_shift=p_shift)
np.random.shuffle(p)
flush_event_id = 0
flush_time = 600
image_features = []
n = len(patches)
i = 1
avg_time = 0
cum_time = 0
dsigma = np.array([0.25, 0.125, 0.0625])
si = CamelyonSlideImage(patches[0][3], None)
is_camelyon = True
area_low = 100
if patches[0][3].find('.ndpi') > -1:
si = NDPISlideImage(patches[0][3], None)
is_camelyon = False
area_low = 50
show_intermediate = False
if 'show' in options:
show_intermediate = True
si._visualize = True
st_elem = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (3, 3))
st_elem_r1 = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (1, 1))
for patch_descriptor in patches:
p_location = patch_descriptor[0]
p_base_size = patch_descriptor[1]
label = patch_descriptor[2]
required=False)
opt_group.add_argument(
"--preview",
type=str,
help='File path for storing the preview image with annotated patches',
default=None,
required=False)
cmdline_args = parser.parse_args()
si = CamelyonSlideImage(image_path=cmdline_args.image,
tissue_mask_path=cmdline_args.mask,
tumor_annotation_file=cmdline_args.tumor_annot)
if '.ndpi' in cmdline_args.image:
si = NDPISlideImage(image_path=cmdline_args.image,
tissue_mask_path=cmdline_args.mask,
tumor_annotation_file=cmdline_args.tumor_annot)
p_shift = cmdline_args.patch_stride
if p_shift < 0:
p_shift = cmdline_args.patch_size
p = si.get_annotated_patches(
extract_level=cmdline_args.extract_level,
min_coverage_extraction=cmdline_args.min_coverage,
min_tumor_coverage=0.6,
p_size=cmdline_args.patch_size,
p_shift=p_shift)
if cmdline_args.preview is not None:
p_img = si.get_patch_visualization(cmdline_args.extract_level,