Downloads: 1
345 lines (344 with data), 9.6 kB
{
"cells": [
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"import numpy as np\n",
"\n",
"import sys\n",
"sys.path.append('..')\n",
"\n",
"import nibabel as nib\n",
"from matplotlib import pyplot as plt\n",
"\n",
"from fetal_net.augment import augment_data"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"def slice_it(arr, inds):\n",
" return arr[inds[0][0]:inds[0][1], inds[1][0]: inds[1][1], inds[2][0]:inds[2][1]]"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"vol = nib.load('../../Datasets/fetus_window_1_99/255/volume.nii')\n",
"mask = nib.load('../../Datasets/fetus_window_1_99/255/truth.nii')\n",
"vol.shape"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"patch_corner = [70, 70, 30]\n",
"patch_shape = [128,128,5]\n",
"data_range = [(start, start + size) for start, size in zip(patch_corner, patch_shape)]\n",
"data_range"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"truth_index = 2\n",
"truth_size = 1\n",
"truth_range = data_range[:2] + [(patch_corner[2] + truth_index,\n",
" patch_corner[2] + truth_index + truth_size)]\n",
"truth_range"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Gaussian Filter"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"data = vol.get_fdata()\n",
"truth = mask.get_fdata()\n",
"data2, truth2, _ = augment_data(data, truth, data.min(), data.max(), data_range=data_range, truth_range=truth_range,\n",
" gaussian_filter={\n",
" 'max_sigma': 1.5,\n",
" 'prob': 1,\n",
" }, poisson_noise=1)\n",
"plt.figure(figsize = (16,14))\n",
"plt.imshow(np.c_[slice_it(data, data_range)[..., 2], data2[..., 2]], cmap='gray')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Shot Noise"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"data = vol.get_fdata()\n",
"truth = mask.get_fdata()\n",
"data2, truth2, _ = augment_data(data, truth, data.min(), data.max(), data_range=data_range, truth_range=truth_range,\n",
" poisson_noise=0.5)\n",
"plt.figure(figsize = (16,14))\n",
"plt.imshow(np.c_[slice_it(data, data_range)[..., 2], data2[..., 2]], cmap='gray')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# contrast deviation"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"data = vol.get_fdata()\n",
"truth = mask.get_fdata()\n",
"print(data.min(), data.max())\n",
"data2, truth2, _ = augment_data(data, truth, data.min(), data.max(), data_range=data_range, truth_range=truth_range,\n",
" contrast_deviation={'min_factor': 0.2, 'max_factor': 0.8})\n",
"plt.figure(figsize = (16,14))\n",
"plt.imshow(np.c_[slice_it(data, data_range)[..., 2], data2[..., 2]], cmap='gray')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# intensity_multiplication_range"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"data = vol.get_fdata()\n",
"truth = mask.get_fdata()\n",
"data2, truth2, _ = augment_data(data, truth, data.min(), data.max(), data_range=data_range, truth_range=truth_range,\n",
" intensity_multiplication_range=[0.8, 1.2])\n",
"plt.figure(figsize = (16,14))\n",
"plt.imshow(np.c_[slice_it(data, data_range)[..., 2], data2[..., 2]], cmap='gray')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# piecewise_affine"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"data = vol.get_fdata()\n",
"truth = mask.get_fdata()\n",
"data2, truth2, _ = augment_data(data, truth, data.min(), data.max(), data_range=data_range, truth_range=truth_range,\n",
" piecewise_affine={'scale': 0.5})\n",
"plt.figure(figsize = (16,14))\n",
"plt.imshow(np.c_[slice_it(data, data_range)[..., 2], data2[..., 2]], cmap='gray')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# elastic_transform"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"data = vol.get_fdata()\n",
"truth = mask.get_fdata()\n",
"data2, truth2, _ = augment_data(data, truth, data.min(), data.max(), data_range=data_range, truth_range=truth_range,\n",
" elastic_transform={'alpha': 5, 'sigma': 1})\n",
"plt.figure(figsize = (16,14))\n",
"plt.imshow(np.c_[slice_it(data, data_range)[..., 2], data2[..., 2]], cmap='gray')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# scale_deviation"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"data = vol.get_fdata()\n",
"truth = mask.get_fdata()\n",
"data2, truth2, _ = augment_data(data, truth, data.min(), data.max(), data_range=data_range, truth_range=truth_range,\n",
" scale_deviation=[0.1, 0.1, 0.0])\n",
"plt.figure(figsize = (16,14))\n",
"plt.imshow(np.c_[slice_it(data, data_range)[..., 2], data2[..., 2]], cmap='gray')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# rotate"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"data = vol.get_fdata()\n",
"truth = mask.get_fdata()\n",
"data2, truth2, _ = augment_data(data, truth, data.min(), data.max(), data_range=data_range, truth_range=truth_range,\n",
" rotate_deviation=[0, 0, 1800])\n",
"plt.figure(figsize = (16,14))\n",
"plt.imshow(np.c_[slice_it(data, data_range)[..., 2], data2[..., 2]], cmap='gray')\n",
"\n",
"#print(slice_it(truth, truth_range).shape)\n",
"#print(truth2.shape)\n",
"plt.figure(figsize = (16,14))\n",
"plt.imshow(np.c_[slice_it(truth, truth_range)[..., 0], truth2[..., 0]], cmap='gray')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Flip"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"data = vol.get_fdata()\n",
"truth = mask.get_fdata()\n",
"data2, truth2, _ = augment_data(data, truth, data.min(), data.max(), data_range=data_range, truth_range=truth_range,\n",
" flip=[0, 0, 1])\n",
"plt.figure(figsize = (16,14))\n",
"plt.imshow(np.c_[slice_it(data, data_range)[..., 2], data2[..., 2]], cmap='gray')\n",
"\n",
"#print(slice_it(truth, truth_range).shape)\n",
"#print(truth2.shape)\n",
"plt.figure(figsize = (16,14))\n",
"plt.imshow(np.c_[slice_it(truth, truth_range)[..., 0], truth2[..., 0]], cmap='gray')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Translate"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"data = vol.get_fdata()\n",
"truth = mask.get_fdata()\n",
"data2, truth2, _ = augment_data(data, truth, data.min(), data.max(), data_range=data_range, truth_range=truth_range,\n",
" translate_deviation=[0, 0, 10])\n",
"plt.figure(figsize = (16,14))\n",
"plt.imshow(np.c_[slice_it(data, data_range)[..., 2], data2[..., 2]], cmap='gray')\n",
"\n",
"#print(slice_it(truth, truth_range).shape)\n",
"#print(truth2.shape)\n",
"plt.figure(figsize = (16,14))\n",
"plt.imshow(np.c_[slice_it(truth, truth_range)[..., 0], truth2[..., 0]], cmap='gray')"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"z_trans = 9\n",
"\n",
"data_range2 = data_range.copy()\n",
"data_range2[-1] = np.add(data_range2[-1], z_trans)\n",
"\n",
"plt.figure(figsize = (16,14))\n",
"plt.imshow(np.c_[slice_it(data, data_range2)[..., 2], data2[..., 2]], cmap='gray')\n",
"\n",
"truth_range2 = truth_range\n",
"truth_range2[-1] = np.add(truth_range2[-1], z_trans)\n",
"plt.figure(figsize = (16,14))\n",
"plt.imshow(np.c_[slice_it(truth, truth_range2)[..., 0], truth2[..., 0]], cmap='gray')"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.6.6"
}
},
"nbformat": 4,
"nbformat_minor": 2
}