Baseline-Networks/model.py at main · muregpro/Baseline-Networks · GitHub

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
from tensorflow.keras import layers
from tensorflow import keras

def get_model(moving_image_shape, fixed_image_shape, with_label_inputs=True, up_filters=[64, 128, 256], down_filters=[256, 128, 64, 32]):

    input_moving_image = keras.Input(moving_image_shape)
    input_fixed_image = keras.Input(fixed_image_shape)

    if with_label_inputs:
        input_moving_label = keras.Input(moving_image_shape)
        input_fixed_label = keras.Input(fixed_image_shape)

    concatenate_layer = layers.Concatenate(axis=-1)([input_moving_image, input_fixed_image])

    ### [First half of the network: downsampling inputs] ###

    # Entry block
    x = layers.Conv3D(32, 3, strides=2, padding="same")(concatenate_layer)
    x = layers.BatchNormalization()(x)
    x = layers.Activation("relu")(x)

    previous_block_activation = x  # Set aside residual

    # Blocks 1, 2, 3 are identical apart from the feature depth.
    for filters in up_filters:
        x = layers.Activation("relu")(x)
        x = layers.Conv3D(filters, 3, padding="same")(x)
        x = layers.BatchNormalization()(x)

        x = layers.Activation("relu")(x)
        x = layers.Conv3D(filters, 3, padding="same")(x)
        x = layers.BatchNormalization()(x)

        x = layers.MaxPooling3D(3, strides=2, padding="same")(x)

        # Project residual
        residual = layers.Conv3D(filters, 1, strides=2, padding="same")(
            previous_block_activation
        )
        x = layers.add([x, residual])  # Add back residual
        previous_block_activation = x  # Set aside next residual

    ### [Second half of the network: upsampling inputs] ###

    for filters in down_filters:
        x = layers.Activation("relu")(x)
        x = layers.Conv3DTranspose(filters, 3, padding="same")(x)
        x = layers.BatchNormalization()(x)

        x = layers.Activation("relu")(x)
        x = layers.Conv3DTranspose(filters, 3, padding="same")(x)
        x = layers.BatchNormalization()(x)

        x = layers.UpSampling3D(2)(x)

        # Project residual
        residual = layers.UpSampling3D(2)(previous_block_activation)
        residual = layers.Conv3D(filters, 1, padding="same")(residual)
        x = layers.add([x, residual])  # Add back residual
        previous_block_activation = x  # Set aside next residual

    # Add a per-pixel classification layer
    out_ddf = layers.Conv3D(3, 3, activation="linear", padding="same")(x)

    # Define the model
    if with_label_inputs:
        model = keras.Model(inputs=[input_moving_image, input_fixed_image, input_moving_label, input_fixed_label], outputs=[out_ddf])
    else:
        model = keras.Model(inputs=[input_moving_image, input_fixed_image], outputs=[out_ddf])
    return model