502 lines (501 with data), 10.9 kB

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Lung Lobes Segmentation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline\n",
    "\n",
    "import SimpleITK as sitk\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import scipy as sp \n",
    "import gui\n",
    "import cv2\n",
    "import matplotlib.image as mpimg\n",
    "\n",
    "# from mayavi import mlab\n",
    "from scipy import signal\n",
    "from myshow import myshow, myshow3d\n",
    "from read_data import LoadData\n",
    "from lung_segment import LungSegment\n",
    "from vessel_segment import VesselSegment\n",
    "from mpl_toolkits.mplot3d import Axes3D"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Read data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_path = \"resource/\"\n",
    "img_name = \"s1.mhd\"\n",
    "# img_name = \"lola11-01.mhd\"\n",
    "data = LoadData(data_path, img_name)\n",
    "data.loaddata()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_path = \"resource/\"\n",
    "img_name = \"s1map.mhd\"\n",
    "labelmap = LoadData(data_path, img_name)\n",
    "labelmap.loaddata()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Lung Lobes Segmentation\n",
    "0. Data Preprocesing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "vs = VesselSegment(original=data.image, closing=labelmap.image)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print \"   Shrik the region of lung...\"\n",
    "vs.erosion(lunglabel=[201, 202])\n",
    "\n",
    "print \"   Pricessing Generate lung mask...\"\n",
    "vs.generate_lung_mask(offset = 1024)\n",
    "\n",
    "# Write image...\n",
    "Lung_mask = sitk.GetImageFromArray(vs.img)\n",
    "sitk.WriteImage(Lung_mask, \"Lung_mask.mhd\")\n",
    "\n",
    "print \"   Processing Downsampling...\"\n",
    "vs.downsampling()\n",
    "\n",
    "print \"   Processing Thresholding...\"\n",
    "vs.thresholding(thval=180)\n",
    "\n",
    "print \"   Processing Region Growing...\"\n",
    "vs.max_filter(filter_size=5)\n",
    "\n",
    "# print \"   Processing Filtering...\"\n",
    "# vs.filtering(min_size=500, max_size=1000)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "filtered = sitk.GetImageFromArray(vs.temp_img)\n",
    "sitk.WriteImage(filtered, \"filtered.mhd\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "filtered = sitk.ReadImage(\"filtered.mhd\")\n",
    "filtered = sitk.GetArrayFromImage(filtered)\n",
    "filtered[filtered > 0] = 1\n",
    "binary_filtered = sitk.GetImageFromArray(filtered)\n",
    "sitk.WriteImage(binary_filtered, \"binary_filtered.mhd\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "### Postprocessing for fissure enhancement"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import SimpleITK as sitk\n",
    "from read_data import LoadData\n",
    "import numpy as np\n",
    "import collections\n",
    "\n",
    "# data = LoadData(path=\"fissure_enhancement_cxx/\", name=\"voxel_val_region_growing_3rd.mhd\")\n",
    "data = LoadData(path=\"\", name=\"filtered_rg.mhd\")\n",
    "data.loaddata()\n",
    "image = sitk.GetArrayFromImage(data.image)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "nonzeros = image[image > 0]\n",
    "d = collections.Counter( nonzeros )\n",
    "val_key = []\n",
    "keys = set([])\n",
    "for key, val in d.items():\n",
    "    # if val > 1000:\n",
    "    if val > 5000:\n",
    "        keys.add(key)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "len(keys)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "image[image == 0] = 1\n",
    "\n",
    "# for p in np.nditer(image, op_flags=['readwrite']):\n",
    "#     if p.tolist() in keys:\n",
    "#         p[...] = 0\n",
    "\n",
    "for key in keys:\n",
    "    image[image == key] = 0\n",
    "\n",
    "image[image > 0] = 1\n",
    "image[image == 0] = 255\n",
    "image[image == 1] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "img = sitk.GetImageFromArray(image.astype(np.uint8))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "sitk.WriteImage(img, \"filtered.mhd\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "size = 7\n",
    "closing = sitk.BinaryMorphologicalClosingImageFilter()\n",
    "closing.SetForegroundValue(255)\n",
    "closing.SetKernelRadius(size)\n",
    "img = closing.Execute(img)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "sitk.WriteImage(img, \"fissure_enhancement_cxx/voxel_val_region_growing_closing.mhd\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data = LoadData(path=\"\", name=\"binary_filtered.mhd\")\n",
    "data.loaddata()\n",
    "vessel = sitk.GetArrayFromImage(data.image)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "fissure = sitk.GetArrayFromImage(img)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import copy\n",
    "fissure_vessel = copy.deepcopy(fissure)\n",
    "fissure_vessel[fissure_vessel != 0] = 1\n",
    "fissure_vessel[vessel != 0] = 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "fissure_vessel_itk = sitk.GetImageFromArray(fissure_vessel)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "sitk.WriteImage(fissure_vessel_itk, \"fissure_enhancement_cxx/fissure_vessel.mhd\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Label map"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "lung_mask = LoadData(path=\"\", name=\"Lung_mask.mhd\")\n",
    "lung_mask.loaddata()\n",
    "fissure = LoadData(path=\"fissure_enhancement_cxx/\", name=\"voxel_val_region_growing_closing.mhd\")\n",
    "fissure.loaddata()\n",
    "vessel = LoadData(path=\"\", name=\"binary_filtered.mhd\")\n",
    "vessel.loaddata()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "lung_mask = sitk.GetArrayFromImage(lung_mask.image)\n",
    "fissure = sitk.GetArrayFromImage(fissure.image)\n",
    "vessel = sitk.GetArrayFromImage(vessel.image)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "lung_mask[lung_mask != 0] = 3\n",
    "lung_mask[vessel > 0] = 1\n",
    "lung_mask[fissure > 0] = 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "lung_mask = sitk.GetImageFromArray(lung_mask)\n",
    "sitk.WriteImage(lung_mask, \"label_map.mhd\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "### Fissure Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "result_dismap = LoadData(path=\"fissure_enhancement_cxx/\", name=\"distmap_voxel_val_rg.mhd\")\n",
    "result_dismap.loaddata()\n",
    "result_dismap_nda = sitk.GetArrayFromImage(result_dismap.image)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "gt_dismap = LoadData(path=\"fissure_enhancement_cxx/\", name=\"distance_map_gt_fissure.mhd\")\n",
    "gt_dismap.loaddata()\n",
    "gt_dismap_nda = sitk.GetArrayFromImage(gt_dismap.image)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import copy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "gt_vals = copy.deepcopy(gt_dismap_nda)\n",
    "gt_vals[result_dismap_nda == 0] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "result_vals = copy.deepcopy(result_dismap_nda)\n",
    "result_vals[gt_dismap_nda == 0] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "num_total = float(np.count_nonzero(gt_vals) + np.count_nonzero(result_vals))\n",
    "mean = float(np.sum(gt_vals) + np.sum(result_vals)) / num_total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "mean * 0.73"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}