-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathlungunetmodel.py
126 lines (98 loc) · 4.6 KB
/
lungunetmodel.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
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
"""
deactivate
conda.bat deactivate
LungUNETCPUEnv\Scripts\activate
python lungunetmodel.py
"""
import numpy as np # linear algebra
import os
import cv2
import warnings
warnings.filterwarnings('ignore')
import tensorflow as tf
from keras.models import *
from keras.layers import *
from keras.optimizers import *
from keras import backend as keras
from keras.preprocessing.image import ImageDataGenerator
from keras.callbacks import ModelCheckpoint, LearningRateScheduler
def dice_coef(y_true, y_pred):
y_true_f = keras.flatten(y_true)
y_pred_f = keras.flatten(y_pred)
intersection = keras.sum(y_true_f * y_pred_f)
return (2. * intersection + 1) / (keras.sum(y_true_f) + keras.sum(y_pred_f) + 1)
def dice_coef_loss(y_true, y_pred):
return -dice_coef(y_true, y_pred)
def unet(input_size=(256,256,1)):
inputs = Input(input_size)
conv1 = Conv2D(32, (3, 3), activation='relu', padding='same')(inputs)
conv1 = Conv2D(32, (3, 3), activation='relu', padding='same')(conv1)
pool1 = MaxPooling2D(pool_size=(2, 2))(conv1)
conv2 = Conv2D(64, (3, 3), activation='relu', padding='same')(pool1)
conv2 = Conv2D(64, (3, 3), activation='relu', padding='same')(conv2)
pool2 = MaxPooling2D(pool_size=(2, 2))(conv2)
conv3 = Conv2D(128, (3, 3), activation='relu', padding='same')(pool2)
conv3 = Conv2D(128, (3, 3), activation='relu', padding='same')(conv3)
pool3 = MaxPooling2D(pool_size=(2, 2))(conv3)
conv4 = Conv2D(256, (3, 3), activation='relu', padding='same')(pool3)
conv4 = Conv2D(256, (3, 3), activation='relu', padding='same')(conv4)
pool4 = MaxPooling2D(pool_size=(2, 2))(conv4)
conv5 = Conv2D(512, (3, 3), activation='relu', padding='same')(pool4)
conv5 = Conv2D(512, (3, 3), activation='relu', padding='same')(conv5)
up6 = concatenate([Conv2DTranspose(256, (2, 2), strides=(2, 2), padding='same')(conv5), conv4], axis=3)
conv6 = Conv2D(256, (3, 3), activation='relu', padding='same')(up6)
conv6 = Conv2D(256, (3, 3), activation='relu', padding='same')(conv6)
up7 = concatenate([Conv2DTranspose(128, (2, 2), strides=(2, 2), padding='same')(conv6), conv3], axis=3)
conv7 = Conv2D(128, (3, 3), activation='relu', padding='same')(up7)
conv7 = Conv2D(128, (3, 3), activation='relu', padding='same')(conv7)
up8 = concatenate([Conv2DTranspose(64, (2, 2), strides=(2, 2), padding='same')(conv7), conv2], axis=3)
conv8 = Conv2D(64, (3, 3), activation='relu', padding='same')(up8)
conv8 = Conv2D(64, (3, 3), activation='relu', padding='same')(conv8)
up9 = concatenate([Conv2DTranspose(32, (2, 2), strides=(2, 2), padding='same')(conv8), conv1], axis=3)
conv9 = Conv2D(32, (3, 3), activation='relu', padding='same')(up9)
conv9 = Conv2D(32, (3, 3), activation='relu', padding='same')(conv9)
conv10 = Conv2D(1, (1, 1), activation='sigmoid')(conv9)
return Model(inputs=[inputs], outputs=[conv10])
model = unet(input_size=(512,512,1))
model.compile(optimizer=Adam(lr=1e-5), loss=dice_coef_loss,
metrics=[dice_coef, 'binary_accuracy'])
# model.summary()
weight_path="cxr_reg_weights.best.hdf5"
model_weights_path = os.path.join(ROOT_DIR,"Weights",weight_path)
model.load_weights(model_weights_path)
"""
Shapes that you wish to resize to
"""
Shape_X,Shape_Y=512,512
def read_image(img_path):
image = cv2.imread(img_path)
image = cv2.resize(image,(Shape_Y,Shape_X))
return image
def get_preds(image):
prep_unet_input_img_1 = image.reshape(1,Shape_X,Shape_Y,1)
prep_unet_input_img = (prep_unet_input_img_1-127.0)/127.0
pred_img = model.predict(prep_unet_input_img)
pred_img_preprocessed_1 = np.squeeze(pred_img)
pred_img_preprocessed = (pred_img_preprocessed_1*255>127).astype(np.int8)
res = cv2.bitwise_and(image,image,mask = pred_img_preprocessed)
return res,pred_img_preprocessed
cv2.imwrite(save_path,res)
if __name__ == '__main__':
INP = os.path.join(ROOT_DIR,"Sample_Inputs")
INP_RESHAPED = os.path.join(ROOT_DIR,"Sample_Inputs_Reshaped")
RES = os.path.join(ROOT_DIR,"Sample_Masked_Results")
MASK_PATH = os.path.join(ROOT_DIR,"Sample_Lung_Masks")
create_folders([INP,INP_RESHAPED,RES])
"""
Images Output :
Original Reshaped Image
Superimposed Lungs Segmentation
"""
input_files = os.listdir(INP)
for i,f in enumerate(input_files):
img = read_image(os.path.join(INP,f))
reshaped_img = deepcopy(img)
segmented_output,mask = get_preds(reshaped_img)
cv2.imwrite(os.path.join(INP_RESHAPED,f),reshaped_img )
cv2.imwrite(os.path.join(RES,f),segmented_output)
cv2.imwrite(os.path.join(MASK_PATH ,f),mask)