154 lines
6.2 KiB
Python
Executable File
154 lines
6.2 KiB
Python
Executable File
from inspect import _ParameterKind
|
|
import SimpleITK as sitk
|
|
import tensorflow as tf
|
|
from tensorflow.keras.models import load_model
|
|
from focal_loss import BinaryFocalLoss
|
|
import numpy as np
|
|
import multiprocessing
|
|
from functools import partial
|
|
import os
|
|
from os import path
|
|
from tqdm import tqdm
|
|
import argparse
|
|
|
|
from sfransen.utils_quintin import *
|
|
from sfransen.DWI_exp.helpers import *
|
|
from sfransen.DWI_exp.preprocessing_function import preprocess
|
|
from sfransen.DWI_exp.callbacks import dice_coef
|
|
#from sfransen.FROC.blob_preprocess import *
|
|
from sfransen.FROC.cal_froc_from_np import *
|
|
from sfransen.load_images import load_images_parrallel
|
|
from sfransen.DWI_exp.losses import weighted_binary_cross_entropy
|
|
from umcglib.froc import *
|
|
from umcglib.binarize import dynamic_threshold
|
|
|
|
|
|
def print_p(*args, **kwargs):
|
|
"""
|
|
Shorthand for print(..., flush=True)
|
|
Useful on HPC cluster where output has buffered writes.
|
|
"""
|
|
print(*args, **kwargs, flush=True)
|
|
|
|
|
|
######## CUDA ################
|
|
os.environ["CUDA_VISIBLE_DEVICES"] = "2"
|
|
N_CPUS = 12
|
|
|
|
|
|
DATA_DIR = "./../data/Nijmegen paths/"
|
|
TARGET_SPACING = (0.5, 0.5, 3)
|
|
INPUT_SHAPE = (192, 192, 24, 3)
|
|
IMAGE_SHAPE = INPUT_SHAPE[:3]
|
|
|
|
final_table = {}
|
|
difference = {}
|
|
for fold in range(5):
|
|
|
|
DATA_SPLIT_INDEX = read_yaml_to_dict(f'./../data/Nijmegen paths/train_val_test_idxs_{fold}.yml')
|
|
TEST_INDEX = DATA_SPLIT_INDEX['test_set0']
|
|
|
|
for img_idx in TEST_INDEX:
|
|
|
|
for model in ['b800','b400']:
|
|
|
|
image_paths = {}
|
|
predictions_added = []
|
|
segmentations_added = []
|
|
images = []
|
|
images_list = []
|
|
segmentations = []
|
|
|
|
if model is 'b800':
|
|
MODEL_PATH = f'./../train_output/calc_exp_t2_b1400calc2_adccalc2_{fold}/models/calc_exp_t2_b1400calc2_adccalc2_{fold}.h5'
|
|
# YAML_DIR = f'./../train_output/calc_exp_t2_b1400calc2_adccalc2_{fold}'
|
|
# IMAGE_DIR = f'./../train_output/calc_exp_t2_b1400calc2_adccalc2_{fold}'
|
|
SERIES = ['t2','b1400calc2','adccalc2']
|
|
if model is 'b400':
|
|
MODEL_PATH = f'./../train_output/calc_exp_t2_b1400calc3_adccalc3_{fold}/models/calc_exp_t2_b1400calc3_adccalc3_{fold}.h5'
|
|
SERIES = ['t2','b1400calc3','adccalc3']
|
|
|
|
for s in SERIES:
|
|
with open(path.join(DATA_DIR, f"{s}.txt"), 'r') as f:
|
|
image_paths[s] = [l.strip() for l in f.readlines()]
|
|
with open(path.join(DATA_DIR, f"seg.txt"), 'r') as f:
|
|
seg_paths = [l.strip() for l in f.readlines()]
|
|
num_images = len(seg_paths)
|
|
|
|
pat_id = os.path.basename(os.path.normpath(seg_paths[img_idx]))[:-7]
|
|
|
|
# print_p("pat_idx:",pat_id)
|
|
|
|
# print(image_paths['t2'][])
|
|
# input('check?')
|
|
# Read and preprocess each of the paths for each series, and the segmentations.
|
|
# print('images number',[TEST_INDEX[img_idx]])
|
|
img_s = {f'{s}': sitk.ReadImage(image_paths[s][img_idx], sitk.sitkFloat32) for s in SERIES}
|
|
seg_s = sitk.ReadImage(seg_paths[img_idx], sitk.sitkFloat32)
|
|
img_n, seg_n = preprocess(img_s, seg_s,
|
|
shape=IMAGE_SHAPE, spacing=TARGET_SPACING)
|
|
for seq in img_n:
|
|
images.append(img_n[f'{seq}'])
|
|
images_list.append(images)
|
|
images = []
|
|
segmentations.append(seg_n)
|
|
|
|
images_list = np.transpose(images_list, (0, 2, 3, 4, 1))
|
|
|
|
########### load module ##################
|
|
# print(' >>>>>>> LOAD MODEL <<<<<<<<<')
|
|
|
|
dependencies = {
|
|
'dice_coef': dice_coef,
|
|
'weighted_cross_entropy_fn':weighted_binary_cross_entropy
|
|
}
|
|
reconstructed_model = load_model(MODEL_PATH, custom_objects=dependencies)
|
|
# reconstructed_model.summary(line_length=120)
|
|
|
|
# make predictions on all TEST_INDEX
|
|
# print(' >>>>>>> START prediction <<<<<<<<<')
|
|
predictions_blur = reconstructed_model.predict(images_list, batch_size=1)
|
|
|
|
############# preprocess #################
|
|
# preprocess predictions by removing the blur and making individual blobs
|
|
# print('>>>>>>>> START preprocess')
|
|
def move_dims(arr):
|
|
# UMCG numpy dimensions convention: dims = (batch, width, heigth, depth)
|
|
# Joeran numpy dimensions convention: dims = (batch, depth, heigth, width)
|
|
arr = np.moveaxis(arr, 3, 1)
|
|
arr = np.moveaxis(arr, 3, 2)
|
|
return arr
|
|
|
|
# Joeran has his numpy arrays ordered differently.
|
|
predictions_blur = move_dims(np.squeeze(predictions_blur,axis=4))
|
|
segmentations = move_dims(segmentations)
|
|
# predictions = [preprocess_softmax(pred, threshold="dynamic")[0] for pred in predictions_blur]
|
|
predictions = predictions_blur
|
|
# print("the size of predictions is:",np.shape(predictions))
|
|
# Remove outer edges
|
|
zeros = np.zeros(np.shape(predictions))
|
|
test = predictions[:,2:-2,2:190,2:190]
|
|
zeros[:,2:-2,2:190,2:190] = test
|
|
predictions = zeros
|
|
|
|
#make list of worst patient predictions
|
|
|
|
|
|
if model is 'b800':
|
|
final_table[pat_id] = [np.max(predictions)]
|
|
print_p(f'Max prediction of {pat_id} in b800 is {np.max(predictions)}')
|
|
if model is 'b400':
|
|
final_table[pat_id].append(np.max(predictions))
|
|
print_p(f'Max prediction of {pat_id} in b400 is {np.max(predictions)}')
|
|
|
|
difference[pat_id] = abs(np.diff(final_table[pat_id]))
|
|
sorted_difference = {k: v for k, v in sorted(difference.items(), key=lambda item: item[1])}
|
|
print_p(f'>>{fold}<<',sorted_difference)
|
|
|
|
|
|
sorted_differences = {k: v for k, v in sorted(difference.items(), key=lambda item: item[1])}
|
|
print_p('>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<')
|
|
print_p(sorted_differences)
|
|
print_p(sorted_differences[::-1])
|
|
|