No description has been provided for this image

W2S dataset¶

In [1]:

  Copied!     
 
# Imports necessary to execute the code
from pathlib import Path

import matplotlib.pyplot as plt
import numpy as np
import pooch
import tifffile
from careamics import CAREamist
from careamics.config import create_n2v_configuration

# use n2v2
use_n2v2 = False

# folder in which to save all the data
root = Path("w2s")
# Imports necessary to execute the code from pathlib import Path import matplotlib.pyplot as plt import numpy as np import pooch import tifffile from careamics import CAREamist from careamics.config import create_n2v_configuration # use n2v2 use_n2v2 = False # folder in which to save all the data root = Path("w2s")

Import the dataset¶

In [ ]:

remove_output

  Copied!     
 
# download the data using pooch
data_root = root / "data"
dataset_url = "https://zenodo.org/records/10925783/files/noisy.tiff?download=1"

file = pooch.retrieve(
    url=dataset_url,
    known_hash="b5cb1dbcb86ce72b8d6e0268498712974b9f38d2fc9e18e8a66673a34aa84215",
    path=data_root,
)
# download the data using pooch data_root = root / "data" dataset_url = "https://zenodo.org/records/10925783/files/noisy.tiff?download=1" file = pooch.retrieve( url=dataset_url, known_hash="b5cb1dbcb86ce72b8d6e0268498712974b9f38d2fc9e18e8a66673a34aa84215", path=data_root, )

Visualize data¶

In [3]:

  Copied!     
 
# load training and validation image and show them side by side
train_image = tifffile.imread(file)
print(f"Image shape: {train_image.shape}")

fig, ax = plt.subplots(1, 3, figsize=(15, 5))
ax[0].imshow(train_image[0][0], cmap="gray")
ax[0].set_title("Channel 1")
ax[1].imshow(train_image[0][1], cmap="gray")
ax[1].set_title("Channel 2")
ax[2].imshow(train_image[0][2], cmap="gray")
ax[2].set_title("Channel 3")
# load training and validation image and show them side by side train_image = tifffile.imread(file) print(f"Image shape: {train_image.shape}") fig, ax = plt.subplots(1, 3, figsize=(15, 5)) ax[0].imshow(train_image[0][0], cmap="gray") ax[0].set_title("Channel 1") ax[1].imshow(train_image[0][1], cmap="gray") ax[1].set_title("Channel 2") ax[2].imshow(train_image[0][2], cmap="gray") ax[2].set_title("Channel 3")

Image shape: (120, 3, 512, 512)

Out[3]:

Text(0.5, 1.0, 'Channel 3')

Train with CAREamics¶

Create configuration¶

In [4]:

  Copied!     
 
# create configuration
algo = "n2v2" if use_n2v2 else "n2v"

config = create_n2v_configuration(
    experiment_name="w2s_" + algo,
    data_type="array",
    axes="SCYX",
    patch_size=(64, 64),
    batch_size=32,
    num_epochs=15,
    n_channels=3,
    use_n2v2=use_n2v2,
)

print(config)
# create configuration algo = "n2v2" if use_n2v2 else "n2v" config = create_n2v_configuration( experiment_name="w2s_" + algo, data_type="array", axes="SCYX", patch_size=(64, 64), batch_size=32, num_epochs=15, n_channels=3, use_n2v2=use_n2v2, ) print(config)

{'algorithm_config': {'algorithm': 'n2v',
                      'loss': 'n2v',
                      'lr_scheduler': {'name': 'ReduceLROnPlateau',
                                       'parameters': {}},
                      'model': {'architecture': 'UNet',
                                'conv_dims': 2,
                                'depth': 2,
                                'final_activation': 'None',
                                'in_channels': 3,
                                'independent_channels': True,
                                'n2v2': False,
                                'num_channels_init': 32,
                                'num_classes': 3},
                      'optimizer': {'name': 'Adam',
                                    'parameters': {'lr': 0.0001}}},
 'data_config': {'axes': 'SCYX',
                 'batch_size': 32,
                 'data_type': 'array',
                 'patch_size': [64, 64],
                 'transforms': [{'flip_x': True,
                                 'flip_y': True,
                                 'name': 'XYFlip',
                                 'p': 0.5},
                                {'name': 'XYRandomRotate90', 'p': 0.5},
                                {'masked_pixel_percentage': 0.2,
                                 'name': 'N2VManipulate',
                                 'roi_size': 11,
                                 'strategy': 'uniform',
                                 'struct_mask_axis': 'none',
                                 'struct_mask_span': 5}]},
 'experiment_name': 'w2s_n2v',
 'training_config': {'checkpoint_callback': {'auto_insert_metric_name': False,
                                             'mode': 'min',
                                             'monitor': 'val_loss',
                                             'save_last': True,
                                             'save_top_k': 3,
                                             'save_weights_only': False,
                                             'verbose': False},
                     'num_epochs': 15},
 'version': '0.1.0'}

Train¶

In [ ]:

remove_output

  Copied!     
 
# instantiate a CAREamist
careamist = CAREamist(
    source=config,
    work_dir=root / algo,
)

# train
careamist.train(
    train_source=train_image,
    val_percentage=0.0,
    val_minimum_split=10,  # use 10 patches as validation
)
# instantiate a CAREamist careamist = CAREamist( source=config, work_dir=root / algo, ) # train careamist.train( train_source=train_image, val_percentage=0.0, val_minimum_split=10, # use 10 patches as validation )

Predict¶

In [ ]:

remove_output

  Copied!     
 
prediction = careamist.predict(
    source=train_image,
    tile_size=(256, 256),
    tile_overlap=(48, 48),
    batch_size=1,
)
prediction = careamist.predict( source=train_image, tile_size=(256, 256), tile_overlap=(48, 48), batch_size=1, )

Save predictions¶

In [7]:

  Copied!     
 
pred_folder = root / ("results_" + algo)
pred_folder.mkdir(exist_ok=True, parents=True)

final_data = np.concatenate(prediction)
tifffile.imwrite(pred_folder / "prediction.tiff", final_data)
pred_folder = root / ("results_" + algo) pred_folder.mkdir(exist_ok=True, parents=True) final_data = np.concatenate(prediction) tifffile.imwrite(pred_folder / "prediction.tiff", final_data)

Visualize the prediction¶

In [8]:

  Copied!     
 
n = 5

fig, ax = plt.subplots(3 * n, 2, figsize=(10, 3 * n * 5))
for i in range(n):
    row = 3 * i

    # channel 1
    ax[row, 0].imshow(train_image[i, 0], cmap="gray")
    ax[row, 0].set_title("Noisy - Channel 1")
    ax[row, 1].imshow(prediction[i].squeeze()[0], cmap="gray")
    ax[row, 1].set_title("Prediction - Channel 1")

    # channel 2
    ax[row + 1, 0].imshow(train_image[i, 1], cmap="gray")
    ax[row + 1, 0].set_title("Noisy - Channel 2")
    ax[row + 1, 1].imshow(prediction[i].squeeze()[1], cmap="gray")
    ax[row + 1, 1].set_title("Prediction - Channel 2")

    # channel 3
    ax[row + 2, 0].imshow(train_image[i, 2], cmap="gray")
    ax[row + 2, 0].set_title("Noisy - Channel 3")
    ax[row + 2, 1].imshow(prediction[i].squeeze()[2], cmap="gray")
    ax[row + 2, 1].set_title("Prediction - Channel 3")
n = 5 fig, ax = plt.subplots(3 * n, 2, figsize=(10, 3 * n * 5)) for i in range(n): row = 3 * i # channel 1 ax[row, 0].imshow(train_image[i, 0], cmap="gray") ax[row, 0].set_title("Noisy - Channel 1") ax[row, 1].imshow(prediction[i].squeeze()[0], cmap="gray") ax[row, 1].set_title("Prediction - Channel 1") # channel 2 ax[row + 1, 0].imshow(train_image[i, 1], cmap="gray") ax[row + 1, 0].set_title("Noisy - Channel 2") ax[row + 1, 1].imshow(prediction[i].squeeze()[1], cmap="gray") ax[row + 1, 1].set_title("Prediction - Channel 2") # channel 3 ax[row + 2, 0].imshow(train_image[i, 2], cmap="gray") ax[row + 2, 0].set_title("Noisy - Channel 3") ax[row + 2, 1].imshow(prediction[i].squeeze()[2], cmap="gray") ax[row + 2, 1].set_title("Prediction - Channel 3")

Cover¶

In [9]:

  Copied!     
 
# create a cover image
im_idx = 8
cv_image_noisy = train_image[im_idx]
cv_image_pred = prediction[im_idx].squeeze()

# create image
cover = np.zeros((3, 256, 256))
(_, height, width) = cv_image_noisy.shape
assert height > 256
assert width > 256

# get min and max and reshape them so that they can be broadcasted with the images
noise_min = np.array(np.min(cv_image_noisy, axis=(1, 2)))[
    (..., *[np.newaxis] * (cv_image_noisy.ndim - 1))
]
noise_max = np.array(np.max(cv_image_noisy, axis=(1, 2)))[
    (..., *[np.newaxis] * (cv_image_noisy.ndim - 1))
]
pred_min = np.array(np.min(cv_image_pred, axis=(1, 2)))[
    (..., *[np.newaxis] * (cv_image_pred.ndim - 1))
]
pred_max = np.array(np.max(cv_image_pred, axis=(1, 2)))[
    (..., *[np.newaxis] * (cv_image_pred.ndim - 1))
]

# normalize train and prediction per channel
norm_noise = (cv_image_noisy - noise_min) / (noise_max - noise_min)
norm_pred = (cv_image_pred - pred_min) / (pred_max - pred_min)

# fill in halves
cover[:, :, : 256 // 2] = norm_noise[
    :,
    height // 2 - 256 // 2 : height // 2 + 256 // 2,
    width // 2 - 256 // 2 : width // 2,
]
cover[:, :, 256 // 2 :] = norm_pred[
    :,
    height // 2 - 256 // 2 : height // 2 + 256 // 2,
    width // 2 : width // 2 + 256 // 2,
]

# move C axis at the end
cover = np.moveaxis(cover, 0, -1)

# plot the single image
plt.imshow(cover)

# save the image
plt.imsave("W2S_N2V.jpeg", cover)
# create a cover image im_idx = 8 cv_image_noisy = train_image[im_idx] cv_image_pred = prediction[im_idx].squeeze() # create image cover = np.zeros((3, 256, 256)) (_, height, width) = cv_image_noisy.shape assert height > 256 assert width > 256 # get min and max and reshape them so that they can be broadcasted with the images noise_min = np.array(np.min(cv_image_noisy, axis=(1, 2)))[ (..., *[np.newaxis] * (cv_image_noisy.ndim - 1)) ] noise_max = np.array(np.max(cv_image_noisy, axis=(1, 2)))[ (..., *[np.newaxis] * (cv_image_noisy.ndim - 1)) ] pred_min = np.array(np.min(cv_image_pred, axis=(1, 2)))[ (..., *[np.newaxis] * (cv_image_pred.ndim - 1)) ] pred_max = np.array(np.max(cv_image_pred, axis=(1, 2)))[ (..., *[np.newaxis] * (cv_image_pred.ndim - 1)) ] # normalize train and prediction per channel norm_noise = (cv_image_noisy - noise_min) / (noise_max - noise_min) norm_pred = (cv_image_pred - pred_min) / (pred_max - pred_min) # fill in halves cover[:, :, : 256 // 2] = norm_noise[ :, height // 2 - 256 // 2 : height // 2 + 256 // 2, width // 2 - 256 // 2 : width // 2, ] cover[:, :, 256 // 2 :] = norm_pred[ :, height // 2 - 256 // 2 : height // 2 + 256 // 2, width // 2 : width // 2 + 256 // 2, ] # move C axis at the end cover = np.moveaxis(cover, 0, -1) # plot the single image plt.imshow(cover) # save the image plt.imsave("W2S_N2V.jpeg", cover)