Get3Dtraintestdata.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 玖零猴的Demo之BraTS3d网络预处理"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## BraTS2018 训练集"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "from __future__ import print_function, division\n",
    "import numpy as np\n",
    "import SimpleITK as sitk\n",
    "import os\n",
    "\n",
    "flair_name = \"_flair.nii.gz\"\n",
    "t1_name = \"_t1.nii.gz\"\n",
    "t1ce_name = \"_t1ce.nii.gz\"\n",
    "t2_name = \"_t2.nii.gz\"\n",
    "mask_name = \"_seg.nii.gz\"\n",
    "\n",
    "input_bratshgg_path = r\"D:\\Project\\CollegeDesign\\dataset\\2-MICCAI_BraTS_2018\\MICCAI_BraTS_2018_Data_Training\\HGG\"\n",
    "input_bratslgg_path = r\"D:\\Project\\CollegeDesign\\dataset\\2-MICCAI_BraTS_2018\\MICCAI_BraTS_2018_Data_Training\\LGG\"\n",
    "input_bratshgg_path2019 = r\"D:\\Project\\CollegeDesign\\dataset\\BraTs2019\\MICCAI_BraTS_2019_Data_Training\\MICCAI_BraTS_2019_Data_Training\\HGG\"\n",
    "input_bratslgg_path2019 = r\"D:\\Project\\CollegeDesign\\dataset\\BraTs2019\\MICCAI_BraTS_2019_Data_Training\\MICCAI_BraTS_2019_Data_Training\\LGG\"\n",
    "\n",
    "output_trainImage = r\"G:\\Project\\CollegeDesign\\dataset\\BraTs3D\\trainImage\"#r\"D:\\Project\\CollegeDesign\\dataset\\BraTs3D\\trainImage\"\n",
    "output_trainMask = r\"G:\\Project\\CollegeDesign\\dataset\\BraTs3D\\trainMask\"#r\"D:\\Project\\CollegeDesign\\dataset\\BraTs3D\\trainMask\"\n",
    "output_testImage = r\"D:\\Project\\CollegeDesign\\dataset\\BraTs3D\\testImage\"\n",
    "output_testMask = r\"D:\\Project\\CollegeDesign\\dataset\\BraTs3D\\testMask\"\n",
    "\n",
    "BLOCKSIZE = (32, 160, 160) #每个分块的大小"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "bratshgg_path = input_bratshgg_path\n",
    "bratslgg_path = input_bratslgg_path\n",
    "if not os.path.exists(bratshgg_path):\n",
    "    print(\"bratshgg_path 不存在\")\n",
    "if not os.path.exists(bratslgg_path):\n",
    "    print(\"bratshgg_path 不存在\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "bratshgg_path2 = input_bratshgg_path2019\n",
    "bratslgg_path2 = input_bratslgg_path2019\n",
    "if not os.path.exists(bratshgg_path2):\n",
    "    print(\"bratshgg_path2019 不存在\")\n",
    "if not os.path.exists(bratslgg_path2):\n",
    "    print(\"bratshgg_path2019 不存在\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "trainImage 输出目录创建成功\n",
      "trainMask 输出目录创建成功\n"
     ]
    }
   ],
   "source": [
    "trainImage = output_trainImage\n",
    "trainMask = output_trainMask\n",
    "if not os.path.exists(trainImage):\n",
    "    os.makedirs(output_trainImage)\n",
    "    print(\"trainImage 输出目录创建成功\")\n",
    "if not os.path.exists(trainMask):\n",
    "    os.makedirs(trainMask)\n",
    "    print(\"trainMask 输出目录创建成功\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "testImage = output_testImage\n",
    "testMask = output_testMask\n",
    "if not os.path.exists(testImage):\n",
    "    os.makedirs(testImage)\n",
    "    print(\"testImage 输出目录创建成功\")\n",
    "if not os.path.exists(testMask):\n",
    "    os.makedirs(testMask)\n",
    "    print(\"testMask 输出目录创建成功\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def file_name_path(file_dir, dir=True, file=False):\n",
    "    \"\"\"\n",
    "    get root path,sub_dirs,all_sub_files\n",
    "    :param file_dir:\n",
    "    :return: dir or file\n",
    "    \"\"\"\n",
    "    for root, dirs, files in os.walk(file_dir):\n",
    "        if len(dirs) and dir:\n",
    "            print(\"sub_dirs:\", dirs)\n",
    "            return dirs\n",
    "        if len(files) and file:\n",
    "            print(\"files:\", files)\n",
    "            return files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "sub_dirs: ['Brats18_2013_10_1', 'Brats18_2013_11_1', 'Brats18_2013_12_1', 'Brats18_2013_13_1', 'Brats18_2013_14_1', 'Brats18_2013_17_1', 'Brats18_2013_18_1', 'Brats18_2013_19_1', 'Brats18_2013_20_1', 'Brats18_2013_21_1', 'Brats18_2013_22_1', 'Brats18_2013_23_1', 'Brats18_2013_25_1', 'Brats18_2013_26_1', 'Brats18_2013_27_1', 'Brats18_2013_2_1', 'Brats18_2013_3_1', 'Brats18_2013_4_1', 'Brats18_2013_5_1', 'Brats18_2013_7_1', 'Brats18_CBICA_AAB_1', 'Brats18_CBICA_AAG_1', 'Brats18_CBICA_AAL_1', 'Brats18_CBICA_AAP_1', 'Brats18_CBICA_ABB_1', 'Brats18_CBICA_ABE_1', 'Brats18_CBICA_ABM_1', 'Brats18_CBICA_ABN_1', 'Brats18_CBICA_ABO_1', 'Brats18_CBICA_ABY_1', 'Brats18_CBICA_ALN_1', 'Brats18_CBICA_ALU_1', 'Brats18_CBICA_ALX_1', 'Brats18_CBICA_AME_1', 'Brats18_CBICA_AMH_1', 'Brats18_CBICA_ANG_1', 'Brats18_CBICA_ANI_1', 'Brats18_CBICA_ANP_1', 'Brats18_CBICA_ANZ_1', 'Brats18_CBICA_AOD_1', 'Brats18_CBICA_AOH_1', 'Brats18_CBICA_AOO_1', 'Brats18_CBICA_AOP_1', 'Brats18_CBICA_AOZ_1', 'Brats18_CBICA_APR_1', 'Brats18_CBICA_APY_1', 'Brats18_CBICA_APZ_1', 'Brats18_CBICA_AQA_1', 'Brats18_CBICA_AQD_1', 'Brats18_CBICA_AQG_1', 'Brats18_CBICA_AQJ_1', 'Brats18_CBICA_AQN_1', 'Brats18_CBICA_AQO_1', 'Brats18_CBICA_AQP_1', 'Brats18_CBICA_AQQ_1', 'Brats18_CBICA_AQR_1', 'Brats18_CBICA_AQT_1', 'Brats18_CBICA_AQU_1', 'Brats18_CBICA_AQV_1', 'Brats18_CBICA_AQY_1', 'Brats18_CBICA_AQZ_1', 'Brats18_CBICA_ARF_1', 'Brats18_CBICA_ARW_1', 'Brats18_CBICA_ARZ_1', 'Brats18_CBICA_ASA_1', 'Brats18_CBICA_ASE_1', 'Brats18_CBICA_ASG_1', 'Brats18_CBICA_ASH_1', 'Brats18_CBICA_ASK_1', 'Brats18_CBICA_ASN_1', 'Brats18_CBICA_ASO_1', 'Brats18_CBICA_ASU_1', 'Brats18_CBICA_ASV_1', 'Brats18_CBICA_ASW_1', 'Brats18_CBICA_ASY_1', 'Brats18_CBICA_ATB_1', 'Brats18_CBICA_ATD_1', 'Brats18_CBICA_ATF_1', 'Brats18_CBICA_ATP_1', 'Brats18_CBICA_ATV_1', 'Brats18_CBICA_ATX_1', 'Brats18_CBICA_AUN_1', 'Brats18_CBICA_AUQ_1', 'Brats18_CBICA_AUR_1', 'Brats18_CBICA_AVG_1', 'Brats18_CBICA_AVJ_1', 'Brats18_CBICA_AVV_1', 'Brats18_CBICA_AWG_1', 'Brats18_CBICA_AWH_1', 'Brats18_CBICA_AWI_1', 'Brats18_CBICA_AXJ_1', 'Brats18_CBICA_AXL_1', 'Brats18_CBICA_AXM_1', 'Brats18_CBICA_AXN_1', 'Brats18_CBICA_AXO_1', 'Brats18_CBICA_AXQ_1', 'Brats18_CBICA_AXW_1', 'Brats18_CBICA_AYA_1', 'Brats18_CBICA_AYI_1', 'Brats18_CBICA_AYU_1', 'Brats18_CBICA_AYW_1', 'Brats18_CBICA_AZD_1', 'Brats18_CBICA_AZH_1', 'Brats18_CBICA_BFB_1', 'Brats18_CBICA_BFP_1', 'Brats18_CBICA_BHB_1', 'Brats18_CBICA_BHK_1', 'Brats18_CBICA_BHM_1', 'Brats18_TCIA01_131_1', 'Brats18_TCIA01_147_1', 'Brats18_TCIA01_150_1', 'Brats18_TCIA01_180_1', 'Brats18_TCIA01_186_1', 'Brats18_TCIA01_190_1', 'Brats18_TCIA01_201_1', 'Brats18_TCIA01_203_1', 'Brats18_TCIA01_221_1', 'Brats18_TCIA01_231_1', 'Brats18_TCIA01_235_1', 'Brats18_TCIA01_335_1', 'Brats18_TCIA01_378_1', 'Brats18_TCIA01_390_1', 'Brats18_TCIA01_401_1', 'Brats18_TCIA01_411_1', 'Brats18_TCIA01_412_1', 'Brats18_TCIA01_425_1', 'Brats18_TCIA01_429_1', 'Brats18_TCIA01_448_1', 'Brats18_TCIA01_460_1', 'Brats18_TCIA01_499_1', 'Brats18_TCIA02_117_1', 'Brats18_TCIA02_118_1', 'Brats18_TCIA02_135_1', 'Brats18_TCIA02_151_1', 'Brats18_TCIA02_168_1', 'Brats18_TCIA02_171_1', 'Brats18_TCIA02_179_1', 'Brats18_TCIA02_198_1', 'Brats18_TCIA02_208_1', 'Brats18_TCIA02_222_1', 'Brats18_TCIA02_226_1', 'Brats18_TCIA02_274_1', 'Brats18_TCIA02_283_1', 'Brats18_TCIA02_290_1', 'Brats18_TCIA02_300_1', 'Brats18_TCIA02_309_1', 'Brats18_TCIA02_314_1', 'Brats18_TCIA02_321_1', 'Brats18_TCIA02_322_1', 'Brats18_TCIA02_331_1', 'Brats18_TCIA02_368_1', 'Brats18_TCIA02_370_1', 'Brats18_TCIA02_374_1', 'Brats18_TCIA02_377_1', 'Brats18_TCIA02_394_1', 'Brats18_TCIA02_430_1', 'Brats18_TCIA02_455_1', 'Brats18_TCIA02_471_1', 'Brats18_TCIA02_473_1', 'Brats18_TCIA02_491_1', 'Brats18_TCIA02_605_1', 'Brats18_TCIA02_606_1', 'Brats18_TCIA02_607_1', 'Brats18_TCIA02_608_1', 'Brats18_TCIA03_121_1', 'Brats18_TCIA03_133_1', 'Brats18_TCIA03_138_1', 'Brats18_TCIA03_199_1', 'Brats18_TCIA03_257_1', 'Brats18_TCIA03_265_1', 'Brats18_TCIA03_296_1', 'Brats18_TCIA03_338_1', 'Brats18_TCIA03_375_1', 'Brats18_TCIA03_419_1', 'Brats18_TCIA03_474_1', 'Brats18_TCIA03_498_1', 'Brats18_TCIA04_111_1', 'Brats18_TCIA04_149_1', 'Brats18_TCIA04_192_1', 'Brats18_TCIA04_328_1', 'Brats18_TCIA04_343_1', 'Brats18_TCIA04_361_1', 'Brats18_TCIA04_437_1', 'Brats18_TCIA04_479_1', 'Brats18_TCIA05_277_1', 'Brats18_TCIA05_396_1', 'Brats18_TCIA05_444_1', 'Brats18_TCIA05_478_1', 'Brats18_TCIA06_165_1', 'Brats18_TCIA06_184_1', 'Brats18_TCIA06_211_1', 'Brats18_TCIA06_247_1', 'Brats18_TCIA06_332_1', 'Brats18_TCIA06_372_1', 'Brats18_TCIA06_409_1', 'Brats18_TCIA06_603_1', 'Brats18_TCIA08_105_1', 'Brats18_TCIA08_113_1', 'Brats18_TCIA08_162_1', 'Brats18_TCIA08_167_1', 'Brats18_TCIA08_205_1', 'Brats18_TCIA08_218_1', 'Brats18_TCIA08_234_1', 'Brats18_TCIA08_242_1', 'Brats18_TCIA08_278_1', 'Brats18_TCIA08_280_1', 'Brats18_TCIA08_319_1', 'Brats18_TCIA08_406_1', 'Brats18_TCIA08_436_1', 'Brats18_TCIA08_469_1']\n",
      "sub_dirs: ['Brats18_2013_0_1', 'Brats18_2013_15_1', 'Brats18_2013_16_1', 'Brats18_2013_1_1', 'Brats18_2013_24_1', 'Brats18_2013_28_1', 'Brats18_2013_29_1', 'Brats18_2013_6_1', 'Brats18_2013_8_1', 'Brats18_2013_9_1', 'Brats18_TCIA09_141_1', 'Brats18_TCIA09_177_1', 'Brats18_TCIA09_254_1', 'Brats18_TCIA09_255_1', 'Brats18_TCIA09_312_1', 'Brats18_TCIA09_402_1', 'Brats18_TCIA09_428_1', 'Brats18_TCIA09_451_1', 'Brats18_TCIA09_462_1', 'Brats18_TCIA09_493_1', 'Brats18_TCIA09_620_1', 'Brats18_TCIA10_103_1', 'Brats18_TCIA10_109_1', 'Brats18_TCIA10_130_1', 'Brats18_TCIA10_152_1', 'Brats18_TCIA10_175_1', 'Brats18_TCIA10_202_1', 'Brats18_TCIA10_241_1', 'Brats18_TCIA10_261_1', 'Brats18_TCIA10_266_1', 'Brats18_TCIA10_276_1', 'Brats18_TCIA10_282_1', 'Brats18_TCIA10_299_1', 'Brats18_TCIA10_307_1', 'Brats18_TCIA10_310_1', 'Brats18_TCIA10_325_1', 'Brats18_TCIA10_330_1', 'Brats18_TCIA10_346_1', 'Brats18_TCIA10_351_1', 'Brats18_TCIA10_387_1', 'Brats18_TCIA10_393_1', 'Brats18_TCIA10_408_1', 'Brats18_TCIA10_410_1', 'Brats18_TCIA10_413_1', 'Brats18_TCIA10_420_1', 'Brats18_TCIA10_442_1', 'Brats18_TCIA10_449_1', 'Brats18_TCIA10_490_1', 'Brats18_TCIA10_625_1', 'Brats18_TCIA10_628_1', 'Brats18_TCIA10_629_1', 'Brats18_TCIA10_632_1', 'Brats18_TCIA10_637_1', 'Brats18_TCIA10_639_1', 'Brats18_TCIA10_640_1', 'Brats18_TCIA10_644_1', 'Brats18_TCIA12_101_1', 'Brats18_TCIA12_249_1', 'Brats18_TCIA12_298_1', 'Brats18_TCIA12_466_1', 'Brats18_TCIA12_470_1', 'Brats18_TCIA12_480_1', 'Brats18_TCIA13_615_1', 'Brats18_TCIA13_618_1', 'Brats18_TCIA13_621_1', 'Brats18_TCIA13_623_1', 'Brats18_TCIA13_624_1', 'Brats18_TCIA13_630_1', 'Brats18_TCIA13_633_1', 'Brats18_TCIA13_634_1', 'Brats18_TCIA13_642_1', 'Brats18_TCIA13_645_1', 'Brats18_TCIA13_650_1', 'Brats18_TCIA13_653_1', 'Brats18_TCIA13_654_1']\n"
     ]
    }
   ],
   "source": [
    "pathhgg_list = file_name_path(bratshgg_path)\n",
    "pathlgg_list = file_name_path(bratslgg_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def normalize(slice, bottom=99, down=1):\n",
    "    \"\"\"\n",
    "    normalize image with mean and std for regionnonzero,and clip the value into range\n",
    "    :param slice:\n",
    "    :param bottom:\n",
    "    :param down:\n",
    "    :return:\n",
    "    \"\"\"\n",
    "    b = np.percentile(slice, bottom)\n",
    "    t = np.percentile(slice, down)\n",
    "    slice = np.clip(slice, t, b)\n",
    "\n",
    "    image_nonzero = slice[np.nonzero(slice)]\n",
    "    if np.std(slice) == 0 or np.std(image_nonzero) == 0:\n",
    "        return slice\n",
    "    else:\n",
    "        tmp = (slice - np.mean(image_nonzero)) / np.std(image_nonzero)\n",
    "        # since the range of intensities is between 0 and 5000 ,\n",
    "        # the min in the normalized slice corresponds to 0 intensity in unnormalized slice\n",
    "        # the min is replaced with -9 just to keep track of 0 intensities\n",
    "        # so that we can discard those intensities afterwards when sampling random patches\n",
    "        tmp[tmp == tmp.min()] = -9\n",
    "        return tmp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "def crop_ceter(img,croph,cropw):   \n",
    "    #for n_slice in range(img.shape[0]):\n",
    "    height,width = img[0].shape \n",
    "    starth = height//2-(croph//2)\n",
    "    startw = width//2-(cropw//2)        \n",
    "    return img[:,starth:starth+croph,startw:startw+cropw]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 人工手動插入切片小實驗"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "a = np.array([[[1,1],[2,2]],[[3,3],[4,4]],[[5,5],[6,6]]])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "b = np.zeros([2,2])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(3, 2, 2)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(2, 2)"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[1, 1],\n",
       "        [2, 2]],\n",
       "\n",
       "       [[3, 3],\n",
       "        [4, 4]],\n",
       "\n",
       "       [[5, 5],\n",
       "        [6, 6]]])"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "c = np.insert(a,0,b,axis = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "4"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "c = np.insert(c,(c.shape[0]),b,axis = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[[0, 0],\n",
       "        [0, 0]],\n",
       "\n",
       "       [[1, 1],\n",
       "        [2, 2]],\n",
       "\n",
       "       [[3, 3],\n",
       "        [4, 4]],\n",
       "\n",
       "       [[5, 5],\n",
       "        [6, 6]],\n",
       "\n",
       "       [[0, 0],\n",
       "        [0, 0]]])"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## BraTS2018 训练集"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Brats18_2013_10_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_2013_11_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_2013_12_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_2013_13_1\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_2013_14_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_2013_17_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_2013_18_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_2013_19_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_2013_20_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_2013_21_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_2013_22_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_2013_23_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_2013_25_1\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_2013_26_1\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_2013_27_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_2013_2_1\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_2013_3_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_2013_4_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_2013_5_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_2013_7_1\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AAB_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_CBICA_AAG_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "Brats18_CBICA_AAL_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_CBICA_AAP_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_CBICA_ABB_1\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ABE_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ABM_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_CBICA_ABN_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "Brats18_CBICA_ABO_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_CBICA_ABY_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_CBICA_ALN_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_CBICA_ALU_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_CBICA_ALX_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_CBICA_AME_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_CBICA_AMH_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_CBICA_ANG_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ANI_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_CBICA_ANP_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_CBICA_ANZ_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "Brats18_CBICA_AOD_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_CBICA_AOH_1\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AOO_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AOP_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AOZ_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_CBICA_APR_1\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_APY_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_CBICA_APZ_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_CBICA_AQA_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "Brats18_CBICA_AQD_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_CBICA_AQG_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AQJ_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AQN_1\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AQO_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AQP_1\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AQQ_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_CBICA_AQR_1\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_CBICA_AQT_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_CBICA_AQU_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AQV_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_CBICA_AQY_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_CBICA_AQZ_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_CBICA_ARF_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ARW_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "Brats18_CBICA_ARZ_1\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ASA_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ASE_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "Brats18_CBICA_ASG_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ASH_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "Brats18_CBICA_ASK_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ASN_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_CBICA_ASO_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ASU_1\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ASV_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ASW_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_CBICA_ASY_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_CBICA_ATB_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ATD_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ATF_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_CBICA_ATP_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_CBICA_ATV_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_CBICA_ATX_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_CBICA_AUN_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AUQ_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_CBICA_AUR_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_CBICA_AVG_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_CBICA_AVJ_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AVV_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AWG_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_CBICA_AWH_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AWI_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AXJ_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AXL_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AXM_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_CBICA_AXN_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AXO_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_CBICA_AXQ_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AXW_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_CBICA_AYA_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "Brats18_CBICA_AYI_1\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AYU_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_CBICA_AYW_1\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_CBICA_AZD_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_CBICA_AZH_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_CBICA_BFB_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_CBICA_BFP_1\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_BHB_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "Brats18_CBICA_BHK_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_CBICA_BHM_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_TCIA01_131_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA01_147_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_TCIA01_150_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_TCIA01_180_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA01_186_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_TCIA01_190_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_TCIA01_201_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA01_203_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_TCIA01_221_1\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA01_231_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Brats18_TCIA01_235_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_TCIA01_335_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_TCIA01_378_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA01_390_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA01_401_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_TCIA01_411_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "Brats18_TCIA01_412_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA01_425_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "Brats18_TCIA01_429_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA01_448_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_TCIA01_460_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA01_499_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "Brats18_TCIA02_117_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_TCIA02_118_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_TCIA02_135_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA02_151_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA02_168_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA02_171_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_TCIA02_179_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_TCIA02_198_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_TCIA02_208_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_TCIA02_222_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA02_226_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_TCIA02_274_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA02_283_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_TCIA02_290_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_TCIA02_300_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_TCIA02_309_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA02_314_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_TCIA02_321_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_TCIA02_322_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "Brats18_TCIA02_331_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_TCIA02_368_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA02_370_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_TCIA02_374_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_TCIA02_377_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_TCIA02_394_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "Brats18_TCIA02_430_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA02_455_1\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA02_471_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_TCIA02_473_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_TCIA02_491_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_TCIA02_605_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_TCIA02_606_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA02_607_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "Brats18_TCIA02_608_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_TCIA03_121_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_TCIA03_133_1\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA03_138_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_TCIA03_199_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA03_257_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA03_265_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_TCIA03_296_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_TCIA03_338_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "Brats18_TCIA03_375_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_TCIA03_419_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_TCIA03_474_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA03_498_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_TCIA04_111_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_TCIA04_149_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "Brats18_TCIA04_192_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA04_328_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_TCIA04_343_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_TCIA04_361_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_TCIA04_437_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_TCIA04_479_1\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA05_277_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_TCIA05_396_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_TCIA05_444_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_TCIA05_478_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA06_165_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_TCIA06_184_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_TCIA06_211_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_TCIA06_247_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "Brats18_TCIA06_332_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_TCIA06_372_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA06_409_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA06_603_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_TCIA08_105_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_TCIA08_113_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA08_162_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_TCIA08_167_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_TCIA08_205_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_TCIA08_218_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_TCIA08_234_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_TCIA08_242_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_TCIA08_278_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_TCIA08_280_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "Brats18_TCIA08_319_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_TCIA08_406_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_TCIA08_436_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_TCIA08_469_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Done!\n"
     ]
    }
   ],
   "source": [
    "part = \"hgg\"\n",
    "for subsetindex in range(len(pathhgg_list)):\n",
    "    print(pathhgg_list[subsetindex])\n",
    "    # 1、读取数据    \n",
    "    brats_subset_path = bratshgg_path + \"/\" + str(pathhgg_list[subsetindex]) + \"/\"\n",
    "    flair_image = brats_subset_path + str(pathhgg_list[subsetindex]) + flair_name\n",
    "    t1_image = brats_subset_path + str(pathhgg_list[subsetindex]) + t1_name\n",
    "    t1ce_image = brats_subset_path + str(pathhgg_list[subsetindex]) + t1ce_name\n",
    "    t2_image = brats_subset_path + str(pathhgg_list[subsetindex]) + t2_name\n",
    "    mask_image = brats_subset_path + str(pathhgg_list[subsetindex]) + mask_name\n",
    "    flair_src = sitk.ReadImage(flair_image, sitk.sitkInt16)\n",
    "    t1_src = sitk.ReadImage(t1_image, sitk.sitkInt16)\n",
    "    t1ce_src = sitk.ReadImage(t1ce_image, sitk.sitkInt16)\n",
    "    t2_src = sitk.ReadImage(t2_image, sitk.sitkInt16)\n",
    "    mask = sitk.ReadImage(mask_image, sitk.sitkUInt8)\n",
    "    flair_array = sitk.GetArrayFromImage(flair_src)\n",
    "    t1_array = sitk.GetArrayFromImage(t1_src)\n",
    "    t1ce_array = sitk.GetArrayFromImage(t1ce_src)\n",
    "    t2_array = sitk.GetArrayFromImage(t2_src)\n",
    "    mask_array = sitk.GetArrayFromImage(mask)\n",
    "    # 2、人工加入切片\n",
    "    myblackslice = np.zeros([240,240])\n",
    "    flair_array = np.insert(flair_array,0,myblackslice,axis = 0)\n",
    "    flair_array = np.insert(flair_array,0,myblackslice,axis = 0)\n",
    "    flair_array = np.insert(flair_array,0,myblackslice,axis = 0)\n",
    "    flair_array = np.insert(flair_array,flair_array.shape[0],myblackslice,axis = 0)\n",
    "    flair_array = np.insert(flair_array,flair_array.shape[0],myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,0,myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,0,myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,0,myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,t1_array.shape[0],myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,t1_array.shape[0],myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,0,myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,0,myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,0,myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,t1ce_array.shape[0],myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,t1ce_array.shape[0],myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,0,myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,0,myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,0,myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,t2_array.shape[0],myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,t2_array.shape[0],myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,0,myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,0,myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,0,myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,mask_array.shape[0],myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,mask_array.shape[0],myblackslice,axis = 0)\n",
    "    \n",
    "    # 3、对四个模态分别进行标准化 \n",
    "    flair_array_nor = normalize(flair_array)\n",
    "    t1_array_nor = normalize(t1_array)\n",
    "    t1ce_array_nor = normalize(t1ce_array)\n",
    "    t2_array_nor = normalize(t2_array)\n",
    "    \n",
    "    # 4、裁剪\n",
    "    flair_crop = crop_ceter(flair_array_nor,160,160)\n",
    "    t1_crop = crop_ceter(t1_array_nor,160,160)\n",
    "    t1ce_crop = crop_ceter(t1ce_array_nor,160,160)\n",
    "    t2_crop = crop_ceter(t2_array_nor,160,160)\n",
    "    mask_crop = crop_ceter(mask_array,160,160) \n",
    "    # 5、分块处理\n",
    "    patch_block_size = BLOCKSIZE\n",
    "    numberxy = patch_block_size[1]\n",
    "    numberz = 8#patch_block_size[0]\n",
    "    width = np.shape(flair_crop)[1]\n",
    "    height = np.shape(flair_crop)[2]\n",
    "    imagez = np.shape(flair_crop)[0]\n",
    "    block_width = np.array(patch_block_size)[1]\n",
    "    block_height = np.array(patch_block_size)[2]\n",
    "    blockz = np.array(patch_block_size)[0]\n",
    "    stridewidth = (width - block_width) // numberxy\n",
    "    strideheight = (height - block_height) // numberxy\n",
    "    stridez = (imagez - blockz) // numberz\n",
    "    step_width = width - (stridewidth * numberxy + block_width)\n",
    "    step_width = step_width // 2\n",
    "    step_height = height - (strideheight * numberxy + block_height)\n",
    "    step_height = step_height // 2\n",
    "    step_z = imagez - (stridez * numberz + blockz)\n",
    "    step_z = step_z // 2\n",
    "    hr_samples_flair_list = []\n",
    "    hr_samples_t1_list = []\n",
    "    hr_samples_t1ce_list = []\n",
    "    hr_samples_t2_list = []\n",
    "    hr_mask_samples_list = []\n",
    "    patchnum = []\n",
    "    for z in range(step_z, numberz * (stridez + 1) + step_z, numberz):\n",
    "        for x in range(step_width, numberxy * (stridewidth + 1) + step_width, numberxy):\n",
    "            for y in range(step_height, numberxy * (strideheight + 1) + step_height, numberxy):\n",
    "                if np.max(mask_crop[z:z + blockz, x:x + block_width, y:y + block_height]) != 0:\n",
    "                    print(\"切%d\"%z)\n",
    "                    patchnum.append(z)\n",
    "                    hr_samples_flair_list.append(flair_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "                    hr_samples_t1_list.append(t1_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "                    hr_samples_t1ce_list.append(t1ce_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "                    hr_samples_t2_list.append(t2_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "                    hr_mask_samples_list.append(mask_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "    samples_flair = np.array(hr_samples_flair_list).reshape((len(hr_samples_flair_list), blockz, block_width, block_height))\n",
    "    samples_t1 = np.array(hr_samples_t1_list).reshape((len(hr_samples_t1_list), blockz, block_width, block_height))\n",
    "    samples_t1ce = np.array(hr_samples_t1ce_list).reshape((len(hr_samples_t1ce_list), blockz, block_width, block_height))\n",
    "    samples_t2 = np.array(hr_samples_t2_list).reshape((len(hr_samples_t2_list), blockz, block_width, block_height))\n",
    "    mask_samples = np.array(hr_mask_samples_list).reshape((len(hr_mask_samples_list), blockz, block_width, block_height))\n",
    "    samples, imagez, height, width = np.shape(samples_flair)[0], np.shape(samples_flair)[1], \\\n",
    "                                     np.shape(samples_flair)[2], np.shape(samples_flair)[3]\n",
    "    # 5、合并和保存\n",
    "    for j in range(samples):\n",
    "        \"\"\"\n",
    "        merage 4 model image into 4 channel (imagez,width,height,channel)\n",
    "        \"\"\"\n",
    "        fourmodelimagearray = np.zeros((imagez, height, width, 4), np.float)\n",
    "        filepath1 = trainImage + \"\\\\\" + part + \"_\" + pathhgg_list[subsetindex] + \"_\" + str(patchnum[j]) + \".npy\"\n",
    "        filepath = trainMask + \"\\\\\" + part + \"_\" + pathhgg_list[subsetindex] + \"_\" + str(patchnum[j]) + \".npy\"\n",
    "        flairimage = samples_flair[j, :, :, :]\n",
    "        flairimage = flairimage.astype(np.float)\n",
    "        fourmodelimagearray[:, :, :, 0] = flairimage\n",
    "        t1image = samples_t1[j, :, :, :]\n",
    "        t1image = t1image.astype(np.float)\n",
    "        fourmodelimagearray[:, :, :, 1] = t1image\n",
    "        t1ceimage = samples_t1ce[j, :, :, :]\n",
    "        t1ceimage = t1ceimage.astype(np.float)\n",
    "        fourmodelimagearray[:, :, :, 2] = t1ceimage\n",
    "        t2image = samples_t2[j, :, :, :]\n",
    "        t2image = t2image.astype(np.float)\n",
    "        fourmodelimagearray[:, :, :, 3] = t2image\n",
    "        np.save(filepath1, fourmodelimagearray)\n",
    "        \n",
    "        wt_tc_etMaskArray = np.zeros((imagez, height, width, 3), np.uint8)\n",
    "        mask_one_sample = mask_samples[j, :, :, :]\n",
    "        WT_Label = mask_one_sample.copy()\n",
    "        WT_Label[mask_one_sample == 1] = 1.\n",
    "        WT_Label[mask_one_sample == 2] = 1.\n",
    "        WT_Label[mask_one_sample == 4] = 1.\n",
    "        TC_Label = mask_one_sample.copy()\n",
    "        TC_Label[mask_one_sample == 1] = 1.\n",
    "        TC_Label[mask_one_sample == 2] = 0.\n",
    "        TC_Label[mask_one_sample == 4] = 1.\n",
    "        ET_Label = mask_one_sample.copy()\n",
    "        ET_Label[mask_one_sample == 1] = 0.\n",
    "        ET_Label[mask_one_sample == 2] = 0.\n",
    "        ET_Label[mask_one_sample == 4] = 1.\n",
    "        wt_tc_etMaskArray[:, :, :, 0] = WT_Label\n",
    "        wt_tc_etMaskArray[:, :, :, 1] = TC_Label\n",
    "        wt_tc_etMaskArray[:, :, :, 2] = ET_Label\n",
    "        np.save(filepath, wt_tc_etMaskArray)\n",
    "print(\"Done!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Brats18_2013_10_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "Brats18_2013_11_1\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_2013_12_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_2013_13_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "Brats18_2013_14_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_2013_17_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "Brats18_2013_18_1\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_2013_19_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_2013_20_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_2013_21_1\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_2013_22_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_2013_23_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "Brats18_2013_25_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_2013_26_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_2013_27_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_2013_2_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_2013_3_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_2013_4_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_2013_5_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_2013_7_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_CBICA_AAB_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_CBICA_AAG_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AAL_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AAP_1\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ABB_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_CBICA_ABE_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_CBICA_ABM_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_CBICA_ABN_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_CBICA_ABO_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ABY_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_CBICA_ALN_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_CBICA_ALU_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ALX_1\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AME_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "Brats18_CBICA_AMH_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "Brats18_CBICA_ANG_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ANI_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_CBICA_ANP_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_CBICA_ANZ_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AOD_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AOH_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AOO_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "Brats18_CBICA_AOP_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AOZ_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_APR_1\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_APY_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_CBICA_APZ_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AQA_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_CBICA_AQD_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_CBICA_AQG_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_CBICA_AQJ_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_CBICA_AQN_1\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AQO_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AQP_1\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AQQ_1\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AQR_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AQT_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_CBICA_AQU_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "Brats18_CBICA_AQV_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_CBICA_AQY_1\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_AQZ_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_CBICA_ARF_1\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "Brats18_CBICA_ARW_1\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ARZ_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_CBICA_ASA_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "Brats18_CBICA_ASE_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "Brats18_CBICA_ASG_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_CBICA_ASH_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "Brats18_CBICA_ASK_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ASN_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "Brats18_CBICA_ASO_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ASU_1\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "Brats18_CBICA_ASV_1\n",
      "切0\n",
      "切8\n",
      "切16\n",
      "切24\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "Brats18_CBICA_ASW_1\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Brats18_CBICA_ASY_1\n",
      "切32\n",
      "切40\n",
      "切48\n",
      "切56\n",
      "切64\n",
      "切72\n",
      "切80\n",
      "切88\n",
      "切96\n",
      "切104\n",
      "切112\n",
      "切120\n",
      "切128\n",
      "Done!\n"
     ]
    }
   ],
   "source": [
    "part = \"lgg\"\n",
    "for subsetindex in range(len(pathlgg_list)):\n",
    "    print(pathhgg_list[subsetindex])\n",
    "    # 1、读取数据    \n",
    "    brats_subset_path = bratslgg_path + \"/\" + str(pathlgg_list[subsetindex]) + \"/\"\n",
    "    flair_image = brats_subset_path + str(pathlgg_list[subsetindex]) + flair_name\n",
    "    t1_image = brats_subset_path + str(pathlgg_list[subsetindex]) + t1_name\n",
    "    t1ce_image = brats_subset_path + str(pathlgg_list[subsetindex]) + t1ce_name\n",
    "    t2_image = brats_subset_path + str(pathlgg_list[subsetindex]) + t2_name\n",
    "    mask_image = brats_subset_path + str(pathlgg_list[subsetindex]) + mask_name\n",
    "    flair_src = sitk.ReadImage(flair_image, sitk.sitkInt16)\n",
    "    t1_src = sitk.ReadImage(t1_image, sitk.sitkInt16)\n",
    "    t1ce_src = sitk.ReadImage(t1ce_image, sitk.sitkInt16)\n",
    "    t2_src = sitk.ReadImage(t2_image, sitk.sitkInt16)\n",
    "    mask = sitk.ReadImage(mask_image, sitk.sitkUInt8)\n",
    "    flair_array = sitk.GetArrayFromImage(flair_src)\n",
    "    t1_array = sitk.GetArrayFromImage(t1_src)\n",
    "    t1ce_array = sitk.GetArrayFromImage(t1ce_src)\n",
    "    t2_array = sitk.GetArrayFromImage(t2_src)\n",
    "    mask_array = sitk.GetArrayFromImage(mask)\n",
    "    # 2、人工加入切片\n",
    "    myblackslice = np.zeros([240,240])\n",
    "    flair_array = np.insert(flair_array,0,myblackslice,axis = 0)\n",
    "    flair_array = np.insert(flair_array,0,myblackslice,axis = 0)\n",
    "    flair_array = np.insert(flair_array,0,myblackslice,axis = 0)\n",
    "    flair_array = np.insert(flair_array,flair_array.shape[0],myblackslice,axis = 0)\n",
    "    flair_array = np.insert(flair_array,flair_array.shape[0],myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,0,myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,0,myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,0,myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,t1_array.shape[0],myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,t1_array.shape[0],myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,0,myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,0,myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,0,myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,t1ce_array.shape[0],myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,t1ce_array.shape[0],myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,0,myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,0,myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,0,myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,t2_array.shape[0],myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,t2_array.shape[0],myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,0,myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,0,myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,0,myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,mask_array.shape[0],myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,mask_array.shape[0],myblackslice,axis = 0)\n",
    "    # 3、对四个模态分别进行标准化 \n",
    "    flair_array_nor = normalize(flair_array)\n",
    "    t1_array_nor = normalize(t1_array)\n",
    "    t1ce_array_nor = normalize(t1ce_array)\n",
    "    t2_array_nor = normalize(t2_array)\n",
    "    # 4、裁剪\n",
    "    flair_crop = crop_ceter(flair_array_nor,160,160)\n",
    "    t1_crop = crop_ceter(t1_array_nor,160,160)\n",
    "    t1ce_crop = crop_ceter(t1ce_array_nor,160,160)\n",
    "    t2_crop = crop_ceter(t2_array_nor,160,160)\n",
    "    mask_crop = crop_ceter(mask_array,160,160) \n",
    "    # 5、分块处理\n",
    "    patch_block_size = BLOCKSIZE\n",
    "    numberxy = patch_block_size[1]\n",
    "    numberz = 8#patch_block_size[0]\n",
    "    width = np.shape(flair_crop)[1]\n",
    "    height = np.shape(flair_crop)[2]\n",
    "    imagez = np.shape(flair_crop)[0]\n",
    "    block_width = np.array(patch_block_size)[1]\n",
    "    block_height = np.array(patch_block_size)[2]\n",
    "    blockz = np.array(patch_block_size)[0]\n",
    "    stridewidth = (width - block_width) // numberxy\n",
    "    strideheight = (height - block_height) // numberxy\n",
    "    stridez = (imagez - blockz) // numberz\n",
    "    step_width = width - (stridewidth * numberxy + block_width)\n",
    "    step_width = step_width // 2\n",
    "    step_height = height - (strideheight * numberxy + block_height)\n",
    "    step_height = step_height // 2\n",
    "    step_z = imagez - (stridez * numberz + blockz)\n",
    "    step_z = step_z // 2\n",
    "    hr_samples_flair_list = []\n",
    "    hr_samples_t1_list = []\n",
    "    hr_samples_t1ce_list = []\n",
    "    hr_samples_t2_list = []\n",
    "    hr_mask_samples_list = []\n",
    "    patchnum = []\n",
    "    for z in range(step_z, numberz * (stridez + 1) + step_z, numberz):\n",
    "        for x in range(step_width, numberxy * (stridewidth + 1) + step_width, numberxy):\n",
    "            for y in range(step_height, numberxy * (strideheight + 1) + step_height, numberxy):\n",
    "                if np.max(mask_crop[z:z + blockz, x:x + block_width, y:y + block_height]) != 0:\n",
    "                    print(\"切%d\"%z)\n",
    "                    patchnum.append(z)\n",
    "                    hr_samples_flair_list.append(flair_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "                    hr_samples_t1_list.append(t1_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "                    hr_samples_t1ce_list.append(t1ce_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "                    hr_samples_t2_list.append(t2_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "                    hr_mask_samples_list.append(mask_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "    samples_flair = np.array(hr_samples_flair_list).reshape((len(hr_samples_flair_list), blockz, block_width, block_height))\n",
    "    samples_t1 = np.array(hr_samples_t1_list).reshape((len(hr_samples_t1_list), blockz, block_width, block_height))\n",
    "    samples_t1ce = np.array(hr_samples_t1ce_list).reshape((len(hr_samples_t1ce_list), blockz, block_width, block_height))\n",
    "    samples_t2 = np.array(hr_samples_t2_list).reshape((len(hr_samples_t2_list), blockz, block_width, block_height))\n",
    "    mask_samples = np.array(hr_mask_samples_list).reshape((len(hr_mask_samples_list), blockz, block_width, block_height))\n",
    "    samples, imagez, height, width = np.shape(samples_flair)[0], np.shape(samples_flair)[1], \\\n",
    "                                     np.shape(samples_flair)[2], np.shape(samples_flair)[3]\n",
    "    # 5、合并和保存\n",
    "    for j in range(samples):\n",
    "        \"\"\"\n",
    "        merage 4 model image into 4 channel (imagez,width,height,channel)\n",
    "        \"\"\"\n",
    "        fourmodelimagearray = np.zeros((imagez, height, width, 4), np.float)\n",
    "        filepath1 = trainImage + \"\\\\\" + part + \"_\" + pathlgg_list[subsetindex] + \"_\" + str(patchnum[j]) + \".npy\"\n",
    "        filepath = trainMask + \"\\\\\" + part + \"_\" + pathlgg_list[subsetindex] + \"_\" + str(patchnum[j]) + \".npy\"\n",
    "        flairimage = samples_flair[j, :, :, :]\n",
    "        flairimage = flairimage.astype(np.float)\n",
    "        fourmodelimagearray[:, :, :, 0] = flairimage\n",
    "        t1image = samples_t1[j, :, :, :]\n",
    "        t1image = t1image.astype(np.float)\n",
    "        fourmodelimagearray[:, :, :, 1] = t1image\n",
    "        t1ceimage = samples_t1ce[j, :, :, :]\n",
    "        t1ceimage = t1ceimage.astype(np.float)\n",
    "        fourmodelimagearray[:, :, :, 2] = t1ceimage\n",
    "        t2image = samples_t2[j, :, :, :]\n",
    "        t2image = t2image.astype(np.float)\n",
    "        fourmodelimagearray[:, :, :, 3] = t2image\n",
    "        np.save(filepath1, fourmodelimagearray)\n",
    "        \n",
    "        wt_tc_etMaskArray = np.zeros((imagez, height, width, 3), np.uint8)\n",
    "        mask_one_sample = mask_samples[j, :, :, :]\n",
    "        WT_Label = mask_one_sample.copy()\n",
    "        WT_Label[mask_one_sample == 1] = 1.\n",
    "        WT_Label[mask_one_sample == 2] = 1.\n",
    "        WT_Label[mask_one_sample == 4] = 1.\n",
    "        TC_Label = mask_one_sample.copy()\n",
    "        TC_Label[mask_one_sample == 1] = 1.\n",
    "        TC_Label[mask_one_sample == 2] = 0.\n",
    "        TC_Label[mask_one_sample == 4] = 1.\n",
    "        ET_Label = mask_one_sample.copy()\n",
    "        ET_Label[mask_one_sample == 1] = 0.\n",
    "        ET_Label[mask_one_sample == 2] = 0.\n",
    "        ET_Label[mask_one_sample == 4] = 1.\n",
    "        wt_tc_etMaskArray[:, :, :, 0] = WT_Label\n",
    "        wt_tc_etMaskArray[:, :, :, 1] = TC_Label\n",
    "        wt_tc_etMaskArray[:, :, :, 2] = ET_Label\n",
    "        np.save(filepath, wt_tc_etMaskArray)\n",
    "print(\"Done!\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 从BraTS2019多的数据集作为测试集"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "def compare_site_names(one,two):\n",
    "    df1 = pd.read_csv(one, header=None)\n",
    "    df2 = pd.read_csv(two, header=None)\n",
    "    data1 = df1.to_dict(orient='list')[0]\n",
    "    data2 = df2.to_dict(orient='list')[0]\n",
    "    diff1 = set(data1).difference(data2)\n",
    "    diff2 = set(data2).difference(data1)\n",
    "    return list(diff2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "hgg_diff = compare_site_names(\"18hgg.csv\",\"19hgg.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "49"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(hgg_diff)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "lgg_diff = compare_site_names(\"18lgg.csv\",\"19lgg.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(lgg_diff)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "pathhgg_list = []\n",
    "pathlgg_list = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "for idx in range(len(hgg_diff)):\n",
    "    mystr = \"BraTS19\" + hgg_diff[idx]\n",
    "    pathhgg_list.append(mystr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['BraTS19_CBICA_BDK_1',\n",
       " 'BraTS19_CBICA_BHQ_1',\n",
       " 'BraTS19_CBICA_AWV_1',\n",
       " 'BraTS19_CBICA_BCL_1',\n",
       " 'BraTS19_CBICA_BGE_1',\n",
       " 'BraTS19_CBICA_BGX_1',\n",
       " 'BraTS19_CBICA_AWX_1',\n",
       " 'BraTS19_CBICA_BHV_1',\n",
       " 'BraTS19_CBICA_AOC_1',\n",
       " 'BraTS19_CBICA_BGW_1',\n",
       " 'BraTS19_CBICA_BAN_1',\n",
       " 'BraTS19_CBICA_AYC_1',\n",
       " 'BraTS19_CBICA_BEM_1',\n",
       " 'BraTS19_TMC_06290_1',\n",
       " 'BraTS19_CBICA_BGN_1',\n",
       " 'BraTS19_CBICA_AUW_1',\n",
       " 'BraTS19_CBICA_BGG_1',\n",
       " 'BraTS19_TMC_11964_1',\n",
       " 'BraTS19_TMC_15477_1',\n",
       " 'BraTS19_TMC_30014_1',\n",
       " 'BraTS19_CBICA_BCF_1',\n",
       " 'BraTS19_CBICA_ANV_1',\n",
       " 'BraTS19_CBICA_BHZ_1',\n",
       " 'BraTS19_CBICA_BKV_1',\n",
       " 'BraTS19_CBICA_AYG_1',\n",
       " 'BraTS19_TMC_06643_1',\n",
       " 'BraTS19_CBICA_BIC_1',\n",
       " 'BraTS19_CBICA_BGO_1',\n",
       " 'BraTS19_CBICA_AVF_1',\n",
       " 'BraTS19_CBICA_ASF_1',\n",
       " 'BraTS19_CBICA_BGT_1',\n",
       " 'BraTS19_CBICA_ASR_1',\n",
       " 'BraTS19_CBICA_APK_1',\n",
       " 'BraTS19_CBICA_BNR_1',\n",
       " 'BraTS19_CBICA_BBG_1',\n",
       " 'BraTS19_CBICA_AUA_1',\n",
       " 'BraTS19_TMC_12866_1',\n",
       " 'BraTS19_CBICA_BLJ_1',\n",
       " 'BraTS19_CBICA_BAX_1',\n",
       " 'BraTS19_CBICA_BAP_1',\n",
       " 'BraTS19_CBICA_BGR_1',\n",
       " 'BraTS19_CBICA_ATN_1',\n",
       " 'BraTS19_TMC_21360_1',\n",
       " 'BraTS19_CBICA_BJY_1',\n",
       " 'BraTS19_CBICA_AOS_1',\n",
       " 'BraTS19_CBICA_AVT_1',\n",
       " 'BraTS19_CBICA_AVB_1',\n",
       " 'BraTS19_TMC_27374_1',\n",
       " 'BraTS19_CBICA_AUX_1']"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pathhgg_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "for idx in range(len(lgg_diff)):\n",
    "    mystr = \"BraTS19\" + lgg_diff[idx]\n",
    "    pathlgg_list.append(mystr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['BraTS19_TMC_09043_1']"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pathlgg_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BraTS19_CBICA_BDK_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BHQ_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_AWV_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BCL_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BGE_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BGX_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_AWX_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BHV_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_AOC_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BGW_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BAN_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_AYC_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BEM_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_TMC_06290_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BGN_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_AUW_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BGG_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_TMC_11964_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_TMC_15477_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_TMC_30014_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BCF_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_ANV_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BHZ_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BKV_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_AYG_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_TMC_06643_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BIC_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BGO_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_AVF_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_ASF_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BGT_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_ASR_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_APK_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BNR_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BBG_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_AUA_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_TMC_12866_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BLJ_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BAX_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BAP_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BGR_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_ATN_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_TMC_21360_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_BJY_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_AOS_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_AVT_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_AVB_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_TMC_27374_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "BraTS19_CBICA_AUX_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "Done!\n"
     ]
    }
   ],
   "source": [
    "part = \"hgg\"\n",
    "for subsetindex in range(len(pathhgg_list)):\n",
    "    print(pathhgg_list[subsetindex])\n",
    "    # 1、读取数据    \n",
    "    brats_subset_path = bratshgg_path2 + \"/\" + str(pathhgg_list[subsetindex]) + \"/\"\n",
    "    flair_image = brats_subset_path + str(pathhgg_list[subsetindex]) + flair_name\n",
    "    t1_image = brats_subset_path + str(pathhgg_list[subsetindex]) + t1_name\n",
    "    t1ce_image = brats_subset_path + str(pathhgg_list[subsetindex]) + t1ce_name\n",
    "    t2_image = brats_subset_path + str(pathhgg_list[subsetindex]) + t2_name\n",
    "    mask_image = brats_subset_path + str(pathhgg_list[subsetindex]) + mask_name\n",
    "    flair_src = sitk.ReadImage(flair_image, sitk.sitkInt16)\n",
    "    t1_src = sitk.ReadImage(t1_image, sitk.sitkInt16)\n",
    "    t1ce_src = sitk.ReadImage(t1ce_image, sitk.sitkInt16)\n",
    "    t2_src = sitk.ReadImage(t2_image, sitk.sitkInt16)\n",
    "    mask = sitk.ReadImage(mask_image, sitk.sitkUInt8)\n",
    "    flair_array = sitk.GetArrayFromImage(flair_src)\n",
    "    t1_array = sitk.GetArrayFromImage(t1_src)\n",
    "    t1ce_array = sitk.GetArrayFromImage(t1ce_src)\n",
    "    t2_array = sitk.GetArrayFromImage(t2_src)\n",
    "    mask_array = sitk.GetArrayFromImage(mask)\n",
    "    # 2、人工加入切片\n",
    "    myblackslice = np.zeros([240,240])\n",
    "    flair_array = np.insert(flair_array,0,myblackslice,axis = 0)\n",
    "    flair_array = np.insert(flair_array,0,myblackslice,axis = 0)\n",
    "    flair_array = np.insert(flair_array,0,myblackslice,axis = 0)\n",
    "    flair_array = np.insert(flair_array,flair_array.shape[0],myblackslice,axis = 0)\n",
    "    flair_array = np.insert(flair_array,flair_array.shape[0],myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,0,myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,0,myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,0,myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,t1_array.shape[0],myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,t1_array.shape[0],myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,0,myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,0,myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,0,myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,t1ce_array.shape[0],myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,t1ce_array.shape[0],myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,0,myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,0,myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,0,myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,t2_array.shape[0],myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,t2_array.shape[0],myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,0,myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,0,myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,0,myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,mask_array.shape[0],myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,mask_array.shape[0],myblackslice,axis = 0)\n",
    "    # 3、对四个模态分别进行标准化 \n",
    "    flair_array_nor = normalize(flair_array)\n",
    "    t1_array_nor = normalize(t1_array)\n",
    "    t1ce_array_nor = normalize(t1ce_array)\n",
    "    t2_array_nor = normalize(t2_array)\n",
    "    # 4、裁剪\n",
    "    flair_crop = crop_ceter(flair_array_nor,160,160)\n",
    "    t1_crop = crop_ceter(t1_array_nor,160,160)\n",
    "    t1ce_crop = crop_ceter(t1ce_array_nor,160,160)\n",
    "    t2_crop = crop_ceter(t2_array_nor,160,160)\n",
    "    mask_crop = crop_ceter(mask_array,160,160) \n",
    "    # 5、分块处理\n",
    "    patch_block_size = BLOCKSIZE\n",
    "    numberxy = patch_block_size[1]\n",
    "    numberz = patch_block_size[0]\n",
    "    width = np.shape(flair_crop)[1]\n",
    "    height = np.shape(flair_crop)[2]\n",
    "    imagez = np.shape(flair_crop)[0]\n",
    "    block_width = np.array(patch_block_size)[1]\n",
    "    block_height = np.array(patch_block_size)[2]\n",
    "    blockz = np.array(patch_block_size)[0]\n",
    "    stridewidth = (width - block_width) // numberxy\n",
    "    strideheight = (height - block_height) // numberxy\n",
    "    stridez = (imagez - blockz) // numberz\n",
    "    step_width = width - (stridewidth * numberxy + block_width)\n",
    "    step_width = step_width // 2\n",
    "    step_height = height - (strideheight * numberxy + block_height)\n",
    "    step_height = step_height // 2\n",
    "    step_z = imagez - (stridez * numberz + blockz)\n",
    "    step_z = step_z // 2\n",
    "    hr_samples_flair_list = []\n",
    "    hr_samples_t1_list = []\n",
    "    hr_samples_t1ce_list = []\n",
    "    hr_samples_t2_list = []\n",
    "    hr_mask_samples_list = []\n",
    "    patchnum = []\n",
    "    for z in range(step_z, numberz * (stridez + 1) + step_z, numberz):\n",
    "        for x in range(step_width, numberxy * (stridewidth + 1) + step_width, numberxy):\n",
    "            for y in range(step_height, numberxy * (strideheight + 1) + step_height, numberxy):\n",
    "                #if np.max(mask_crop[z:z + blockz, x:x + block_width, y:y + block_height]) != 0:\n",
    "                print(\"切%d\"%z)\n",
    "                patchnum.append(z)\n",
    "                hr_samples_flair_list.append(flair_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "                hr_samples_t1_list.append(t1_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "                hr_samples_t1ce_list.append(t1ce_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "                hr_samples_t2_list.append(t2_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "                hr_mask_samples_list.append(mask_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "    samples_flair = np.array(hr_samples_flair_list).reshape((len(hr_samples_flair_list), blockz, block_width, block_height))\n",
    "    samples_t1 = np.array(hr_samples_t1_list).reshape((len(hr_samples_t1_list), blockz, block_width, block_height))\n",
    "    samples_t1ce = np.array(hr_samples_t1ce_list).reshape((len(hr_samples_t1ce_list), blockz, block_width, block_height))\n",
    "    samples_t2 = np.array(hr_samples_t2_list).reshape((len(hr_samples_t2_list), blockz, block_width, block_height))\n",
    "    mask_samples = np.array(hr_mask_samples_list).reshape((len(hr_mask_samples_list), blockz, block_width, block_height))\n",
    "    samples, imagez, height, width = np.shape(samples_flair)[0], np.shape(samples_flair)[1], \\\n",
    "                                     np.shape(samples_flair)[2], np.shape(samples_flair)[3]\n",
    "    # 5、合并和保存\n",
    "    for j in range(samples):\n",
    "        \"\"\"\n",
    "        merage 4 model image into 4 channel (imagez,width,height,channel)\n",
    "        \"\"\"\n",
    "        fourmodelimagearray = np.zeros((imagez, height, width, 4), np.float)\n",
    "        filepath1 = testImage + \"\\\\\" + part + \"_\" + pathhgg_list[subsetindex] + \"_\" + str(patchnum[j]) + \".npy\"\n",
    "        filepath = testMask + \"\\\\\" + part + \"_\" + pathhgg_list[subsetindex] + \"_\" + str(patchnum[j]) + \".npy\"\n",
    "        flairimage = samples_flair[j, :, :, :]\n",
    "        flairimage = flairimage.astype(np.float)\n",
    "        fourmodelimagearray[:, :, :, 0] = flairimage\n",
    "        t1image = samples_t1[j, :, :, :]\n",
    "        t1image = t1image.astype(np.float)\n",
    "        fourmodelimagearray[:, :, :, 1] = t1image\n",
    "        t1ceimage = samples_t1ce[j, :, :, :]\n",
    "        t1ceimage = t1ceimage.astype(np.float)\n",
    "        fourmodelimagearray[:, :, :, 2] = t1ceimage\n",
    "        t2image = samples_t2[j, :, :, :]\n",
    "        t2image = t2image.astype(np.float)\n",
    "        fourmodelimagearray[:, :, :, 3] = t2image\n",
    "        np.save(filepath1, fourmodelimagearray)\n",
    "        \n",
    "        wt_tc_etMaskArray = np.zeros((imagez, height, width, 3), np.uint8)\n",
    "        mask_one_sample = mask_samples[j, :, :, :]\n",
    "        WT_Label = mask_one_sample.copy()\n",
    "        WT_Label[mask_one_sample == 1] = 1.\n",
    "        WT_Label[mask_one_sample == 2] = 1.\n",
    "        WT_Label[mask_one_sample == 4] = 1.\n",
    "        TC_Label = mask_one_sample.copy()\n",
    "        TC_Label[mask_one_sample == 1] = 1.\n",
    "        TC_Label[mask_one_sample == 2] = 0.\n",
    "        TC_Label[mask_one_sample == 4] = 1.\n",
    "        ET_Label = mask_one_sample.copy()\n",
    "        ET_Label[mask_one_sample == 1] = 0.\n",
    "        ET_Label[mask_one_sample == 2] = 0.\n",
    "        ET_Label[mask_one_sample == 4] = 1.\n",
    "        wt_tc_etMaskArray[:, :, :, 0] = WT_Label\n",
    "        wt_tc_etMaskArray[:, :, :, 1] = TC_Label\n",
    "        wt_tc_etMaskArray[:, :, :, 2] = ET_Label\n",
    "        np.save(filepath, wt_tc_etMaskArray)\n",
    "print(\"Done!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['BraTS19_TMC_09043_1']"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pathlgg_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "BraTS19_TMC_09043_1\n",
      "切0\n",
      "切32\n",
      "切64\n",
      "切96\n",
      "切128\n",
      "Done!\n"
     ]
    }
   ],
   "source": [
    "part = \"lgg\"\n",
    "for subsetindex in range(len(pathlgg_list)):\n",
    "    print(pathlgg_list[subsetindex])\n",
    "    # 1、读取数据    \n",
    "    brats_subset_path = bratslgg_path2 + \"/\" + str(pathlgg_list[subsetindex]) + \"/\"\n",
    "    flair_image = brats_subset_path + str(pathlgg_list[subsetindex]) + flair_name\n",
    "    t1_image = brats_subset_path + str(pathlgg_list[subsetindex]) + t1_name\n",
    "    t1ce_image = brats_subset_path + str(pathlgg_list[subsetindex]) + t1ce_name\n",
    "    t2_image = brats_subset_path + str(pathlgg_list[subsetindex]) + t2_name\n",
    "    mask_image = brats_subset_path + str(pathlgg_list[subsetindex]) + mask_name\n",
    "    flair_src = sitk.ReadImage(flair_image, sitk.sitkInt16)\n",
    "    t1_src = sitk.ReadImage(t1_image, sitk.sitkInt16)\n",
    "    t1ce_src = sitk.ReadImage(t1ce_image, sitk.sitkInt16)\n",
    "    t2_src = sitk.ReadImage(t2_image, sitk.sitkInt16)\n",
    "    mask = sitk.ReadImage(mask_image, sitk.sitkUInt8)\n",
    "    flair_array = sitk.GetArrayFromImage(flair_src)\n",
    "    t1_array = sitk.GetArrayFromImage(t1_src)\n",
    "    t1ce_array = sitk.GetArrayFromImage(t1ce_src)\n",
    "    t2_array = sitk.GetArrayFromImage(t2_src)\n",
    "    mask_array = sitk.GetArrayFromImage(mask)\n",
    "    # 2、人工加入切片\n",
    "    myblackslice = np.zeros([240,240])\n",
    "    flair_array = np.insert(flair_array,0,myblackslice,axis = 0)\n",
    "    flair_array = np.insert(flair_array,0,myblackslice,axis = 0)\n",
    "    flair_array = np.insert(flair_array,0,myblackslice,axis = 0)\n",
    "    flair_array = np.insert(flair_array,flair_array.shape[0],myblackslice,axis = 0)\n",
    "    flair_array = np.insert(flair_array,flair_array.shape[0],myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,0,myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,0,myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,0,myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,t1_array.shape[0],myblackslice,axis = 0)\n",
    "    t1_array = np.insert(t1_array,t1_array.shape[0],myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,0,myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,0,myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,0,myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,t1ce_array.shape[0],myblackslice,axis = 0)\n",
    "    t1ce_array = np.insert(t1ce_array,t1ce_array.shape[0],myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,0,myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,0,myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,0,myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,t2_array.shape[0],myblackslice,axis = 0)\n",
    "    t2_array = np.insert(t2_array,t2_array.shape[0],myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,0,myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,0,myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,0,myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,mask_array.shape[0],myblackslice,axis = 0)\n",
    "    mask_array = np.insert(mask_array,mask_array.shape[0],myblackslice,axis = 0)\n",
    "    # 3、对四个模态分别进行标准化 \n",
    "    flair_array_nor = normalize(flair_array)\n",
    "    t1_array_nor = normalize(t1_array)\n",
    "    t1ce_array_nor = normalize(t1ce_array)\n",
    "    t2_array_nor = normalize(t2_array)\n",
    "    # 4、裁剪\n",
    "    flair_crop = crop_ceter(flair_array_nor,160,160)\n",
    "    t1_crop = crop_ceter(t1_array_nor,160,160)\n",
    "    t1ce_crop = crop_ceter(t1ce_array_nor,160,160)\n",
    "    t2_crop = crop_ceter(t2_array_nor,160,160)\n",
    "    mask_crop = crop_ceter(mask_array,160,160) \n",
    "    # 5、分块处理\n",
    "    patch_block_size = BLOCKSIZE\n",
    "    numberxy = patch_block_size[1]\n",
    "    numberz = patch_block_size[0]\n",
    "    width = np.shape(flair_crop)[1]\n",
    "    height = np.shape(flair_crop)[2]\n",
    "    imagez = np.shape(flair_crop)[0]\n",
    "    block_width = np.array(patch_block_size)[1]\n",
    "    block_height = np.array(patch_block_size)[2]\n",
    "    blockz = np.array(patch_block_size)[0]\n",
    "    stridewidth = (width - block_width) // numberxy\n",
    "    strideheight = (height - block_height) // numberxy\n",
    "    stridez = (imagez - blockz) // numberz\n",
    "    step_width = width - (stridewidth * numberxy + block_width)\n",
    "    step_width = step_width // 2\n",
    "    step_height = height - (strideheight * numberxy + block_height)\n",
    "    step_height = step_height // 2\n",
    "    step_z = imagez - (stridez * numberz + blockz)\n",
    "    step_z = step_z // 2\n",
    "    hr_samples_flair_list = []\n",
    "    hr_samples_t1_list = []\n",
    "    hr_samples_t1ce_list = []\n",
    "    hr_samples_t2_list = []\n",
    "    hr_mask_samples_list = []\n",
    "    patchnum = []\n",
    "    for z in range(step_z, numberz * (stridez + 1) + step_z, numberz):\n",
    "        for x in range(step_width, numberxy * (stridewidth + 1) + step_width, numberxy):\n",
    "            for y in range(step_height, numberxy * (strideheight + 1) + step_height, numberxy):\n",
    "                #if np.max(mask_crop[z:z + blockz, x:x + block_width, y:y + block_height]) != 0:\n",
    "                print(\"切%d\"%z)\n",
    "                patchnum.append(z)\n",
    "                hr_samples_flair_list.append(flair_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "                hr_samples_t1_list.append(t1_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "                hr_samples_t1ce_list.append(t1ce_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "                hr_samples_t2_list.append(t2_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "                hr_mask_samples_list.append(mask_crop[z:z + blockz, x:x + block_width, y:y + block_height])\n",
    "    samples_flair = np.array(hr_samples_flair_list).reshape((len(hr_samples_flair_list), blockz, block_width, block_height))\n",
    "    samples_t1 = np.array(hr_samples_t1_list).reshape((len(hr_samples_t1_list), blockz, block_width, block_height))\n",
    "    samples_t1ce = np.array(hr_samples_t1ce_list).reshape((len(hr_samples_t1ce_list), blockz, block_width, block_height))\n",
    "    samples_t2 = np.array(hr_samples_t2_list).reshape((len(hr_samples_t2_list), blockz, block_width, block_height))\n",
    "    mask_samples = np.array(hr_mask_samples_list).reshape((len(hr_mask_samples_list), blockz, block_width, block_height))\n",
    "    samples, imagez, height, width = np.shape(samples_flair)[0], np.shape(samples_flair)[1], \\\n",
    "                                     np.shape(samples_flair)[2], np.shape(samples_flair)[3]\n",
    "    # 5、合并和保存\n",
    "    for j in range(samples):\n",
    "        \"\"\"\n",
    "        merage 4 model image into 4 channel (imagez,width,height,channel)\n",
    "        \"\"\"\n",
    "        fourmodelimagearray = np.zeros((imagez, height, width, 4), np.float)\n",
    "        filepath1 = testImage + \"\\\\\" + part + \"_\" + pathlgg_list[subsetindex] + \"_\" + str(patchnum[j]) + \".npy\"\n",
    "        filepath = testMask + \"\\\\\" + part + \"_\" + pathlgg_list[subsetindex] + \"_\" + str(patchnum[j]) + \".npy\"\n",
    "        flairimage = samples_flair[j, :, :, :]\n",
    "        flairimage = flairimage.astype(np.float)\n",
    "        fourmodelimagearray[:, :, :, 0] = flairimage\n",
    "        t1image = samples_t1[j, :, :, :]\n",
    "        t1image = t1image.astype(np.float)\n",
    "        fourmodelimagearray[:, :, :, 1] = t1image\n",
    "        t1ceimage = samples_t1ce[j, :, :, :]\n",
    "        t1ceimage = t1ceimage.astype(np.float)\n",
    "        fourmodelimagearray[:, :, :, 2] = t1ceimage\n",
    "        t2image = samples_t2[j, :, :, :]\n",
    "        t2image = t2image.astype(np.float)\n",
    "        fourmodelimagearray[:, :, :, 3] = t2image\n",
    "        np.save(filepath1, fourmodelimagearray)\n",
    "        \n",
    "        wt_tc_etMaskArray = np.zeros((imagez, height, width, 3), np.uint8)\n",
    "        mask_one_sample = mask_samples[j, :, :, :]\n",
    "        WT_Label = mask_one_sample.copy()\n",
    "        WT_Label[mask_one_sample == 1] = 1.\n",
    "        WT_Label[mask_one_sample == 2] = 1.\n",
    "        WT_Label[mask_one_sample == 4] = 1.\n",
    "        TC_Label = mask_one_sample.copy()\n",
    "        TC_Label[mask_one_sample == 1] = 1.\n",
    "        TC_Label[mask_one_sample == 2] = 0.\n",
    "        TC_Label[mask_one_sample == 4] = 1.\n",
    "        ET_Label = mask_one_sample.copy()\n",
    "        ET_Label[mask_one_sample == 1] = 0.\n",
    "        ET_Label[mask_one_sample == 2] = 0.\n",
    "        ET_Label[mask_one_sample == 4] = 1.\n",
    "        wt_tc_etMaskArray[:, :, :, 0] = WT_Label\n",
    "        wt_tc_etMaskArray[:, :, :, 1] = TC_Label\n",
    "        wt_tc_etMaskArray[:, :, :, 2] = ET_Label\n",
    "        np.save(filepath, wt_tc_etMaskArray)\n",
    "print(\"Done!\")"
   ]
  },
  {
   "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": "3.6.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}