forked from yankai317/segment
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathval.py
104 lines (80 loc) · 4.03 KB
/
val.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
from get_data import *
import numpy as np
import random, re
from statistics import *
import time
from skimage import measure
def net_val(model, root, weights, times, down_sample): # 执行评估步骤
avg_liver = []
C_SIZE = model.batch_size
W_SIZE = model.width
H_SIZE = model.height
THRESHOLD = model.threshold
file_list = [file for file in os.listdir(root) if
int(re.sub("\D", "", file)) >= 30 and int(re.sub("\D", "", file)) <= 130 and "volume" in file]
for i, file in enumerate(file_list):
t0 = time.time()
data_array, label_array = read_data(root, file, 1, down_sample)
out_vtk_liver = np.zeros(data_array.shape)
indexs = chop_datas(data_array, W_SIZE, H_SIZE, C_SIZE, train=False) # 获取到每个长方体的起始点的坐标索引
data_list = list(indexs)
for j, index in enumerate(data_list): # 遍历索引，并将长方体分别送入网络执行训练
c_start, c_end, w_start, w_end, h_start, h_end = (index[k] for k in range(6))
data = data_array[c_start:c_end, w_start:w_end, h_start:h_end]
label = label_array[c_start:c_end, w_start:w_end, h_start:h_end]
loss, liver, liver_iou, learn_rate, step = model._val(data, label, weights)
out_vtk_liver[c_start:c_end, w_start:w_end, h_start:h_end] = liver
over = (i == len(file_list) - 1)
ious = []
for thr in [0.4, 0.5, 0.6]:
out_vtk_liver_with_thr = 1.0 * (out_vtk_liver > thr)
label_array_obj = 1.0 * (label_array[:, :, :, 0] > thr)
total_iou = np.sum(1.0 * out_vtk_liver_with_thr * label_array_obj) / (
1.0 * np.sum((out_vtk_liver_with_thr + label_array_obj) > 0))
ious.append(total_iou)
add_val(times, model.save_path, file, ious, over)
out_vtk_liver = 1.0 * (out_vtk_liver > THRESHOLD)
label_array_obj = 1.0 * (label_array[:, :, :, 0] > THRESHOLD)
# post process
# pred_seg = measure.label(out_vtk_liver, 4)
# props = measure.regionprops(pred_seg)
#
# max_area = 0
# max_index = 0
# for index, prop in enumerate(props, start=1):
# if prop.area > max_area:
# max_area = prop.area
# max_index = index
#
# pred_seg[pred_seg != max_index] = 0
# pred_seg[pred_seg == max_index] = 1
#
# out_vtk_liver = pred_seg.astype(np.uint8)
total_dice = np.sum(2.0 * out_vtk_liver * label_array_obj) / (1.0 * np.sum(out_vtk_liver + label_array_obj))
total_iou = np.sum(1.0 * out_vtk_liver * label_array_obj) / (
1.0 * np.sum((out_vtk_liver + label_array_obj) > 0))
print("before: ", ious, "after: ", total_iou)
model._run_accurary(total_iou) # 将整体准确度上传tensorboard
print("=================================================================================")
print("=================================================================================")
print("=================================================================================")
print(
"time: " + str(times) + " test: " + str(i + 1) + " total_dice: " + str(total_dice),
" total_iou: " + str(total_iou),
)
t = time.time() - t0
print("cost time:", t)
print("\n\n\n")
if os.path.exists(os.path.join(model.save_path, 'vtk')):
pass
else:
os.makedirs(os.path.join(model.save_path, 'vtk'))
origin_dicom = sitk.ReadImage(os.path.join(root, file))
seg_result = sitk.GetImageFromArray(out_vtk_liver)
seg_result.SetDirection(origin_dicom.GetDirection())
seg_result.SetOrigin(origin_dicom.GetOrigin())
seg_result.SetSpacing(origin_dicom.GetSpacing())
sitk.WriteImage(seg_result, os.path.join(os.path.join(model.save_path, 'vtk'), file))
avg_liver.append(total_iou)
del out_vtk_liver,seg_result, origin_dicom
return avg_liver