Corrected bug where masks were not saved in the right location and he…

…nce the following nodes were wrongly working

notebooks/brainExtractionU-Net.ipynb

@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 29,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -45,7 +45,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 30,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -60,7 +60,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 45,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -84,6 +84,10 @@
     "    \n",
     "    def _run_interface(self, runtime): \n",
     "        \n",
+    "        #_, name, ext = split_filename(os.path.abspath(self.inputs.in_file))\n",
+    "        #out_file = os.path.join(os.getcwd().replace(self.inputs.base_dir,'/fetaldata'), ''.join((name, self.inputs.out_postfix, ext)))\n",
+    "        print(\"9lawi:\")\n",
+    "        #print(out_file)\n",
     "        cmd = 'python brainExtractionTwoSteps.py -i \"{}\" -c \"{}\" -t \"{}\" -C \"{}\" -T \"{}\" -o \"{}\" '.format(self.inputs.in_file,self.inputs.in_ckpt_loc,\n",
     "                                                                         self.inputs.threshold_loc,self.inputs.in_ckpt_seg,\n",
     "                                                                         self.inputs.threshold_seg,self.inputs.out_postfix)\n",
@@ -96,9 +100,11 @@
     "        return runtime\n",
     "\n",
     "    def _list_outputs(self):\n",
-    "\n",
+    "        #outputs = self._outputs().get()\n",
+    "        #_, name, ext = split_filename(os.path.abspath(self.inputs.in_file))\n",
+    "        #outputs['out_file'] = os.path.join(os.getcwd().replace(self.inputs.base_dir,'/fetaldata'), ''.join((name, self.inputs.out_postfix, ext)))\n",
     "        return {'out_file': self.inputs.in_file[:-4]+self.inputs.out_postfix}\n",
-    "    \n",
+    "        #return outputs\n",
     "    \n",
     "class MultipleBrainExtractionInputSpec(BaseInterfaceInputSpec):\n",
     "    base_dir = Directory(desc='Root directory',mandatory=True,exists=True)\n",
@@ -131,31 +137,34 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 28,
+   "execution_count": 46,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "200819-18:26:32,370 nipype.workflow INFO:\n",
+      "200826-16:58:06,787 nipype.workflow INFO:\n",
       "\t Workflow BrainExtractTwoStepsWorfklow settings: ['check', 'execution', 'logging', 'monitoring']\n",
-      "200819-18:26:32,374 nipype.workflow INFO:\n",
+      "200826-16:58:06,791 nipype.workflow INFO:\n",
       "\t Running serially.\n",
-      "200819-18:26:32,375 nipype.workflow INFO:\n",
+      "200826-16:58:06,792 nipype.workflow INFO:\n",
       "\t [Node] Setting-up \"BrainExtractTwoStepsWorfklow.data_grabber\" in \"/media/hkebiri/Storage/dti_t2_ANDRAS/export_normal/original-t2/NIFTI_HK/TestBrainExtraction/BrainExtractTwoStepsWorfklow/data_grabber\".\n",
-      "200819-18:26:32,379 nipype.workflow INFO:\n",
+      "200826-16:58:06,796 nipype.workflow INFO:\n",
       "\t [Node] Running \"data_grabber\" (\"nipype.interfaces.io.DataGrabber\")\n",
-      "200819-18:26:32,383 nipype.workflow INFO:\n",
+      "200826-16:58:06,799 nipype.workflow INFO:\n",
       "\t [Node] Finished \"BrainExtractTwoStepsWorfklow.data_grabber\".\n",
-      "200819-18:26:32,384 nipype.workflow INFO:\n",
+      "200826-16:58:06,800 nipype.workflow INFO:\n",
       "\t [Node] Setting-up \"BrainExtractTwoStepsWorfklow.Multiple_Brain_extraction\" in \"/media/hkebiri/Storage/dti_t2_ANDRAS/export_normal/original-t2/NIFTI_HK/TestBrainExtraction/BrainExtractTwoStepsWorfklow/Multiple_Brain_extraction\".\n",
-      "200819-18:26:32,391 nipype.workflow INFO:\n",
+      "200826-16:58:06,803 nipype.workflow INFO:\n",
       "\t [Node] Running \"Multiple_Brain_extraction\" (\"__main__.MultipleBrainExtraction\")\n",
+      "9lawi:\n",
       "... cmd: python brainExtractionTwoSteps.py -i \"/media/hkebiri/Storage/dti_t2_ANDRAS/export_normal/original-t2/NIFTI_HK/TestBrainExtraction/6_twoSteps/Image6_worldmatrixModified_lr.nii\" -c \"/home/hkebiri/Desktop/mialsuperresolutiontoolkit/pymialsrtk/data/Network_checkpoints_localization/Unet.ckpt-88000\" -t \"0.49\" -C \"/home/hkebiri/Desktop/mialsuperresolutiontoolkit/pymialsrtk/data/Network_checkpoints_segmentation/Unet.ckpt-20000\" -T \"0.5\" -o \"_masked.nii.gz\" \n",
+      "9lawi:\n",
       "... cmd: python brainExtractionTwoSteps.py -i \"/media/hkebiri/Storage/dti_t2_ANDRAS/export_normal/original-t2/NIFTI_HK/TestBrainExtraction/6_twoSteps/Image0_worldmatrixModified_lr.nii\" -c \"/home/hkebiri/Desktop/mialsuperresolutiontoolkit/pymialsrtk/data/Network_checkpoints_localization/Unet.ckpt-88000\" -t \"0.49\" -C \"/home/hkebiri/Desktop/mialsuperresolutiontoolkit/pymialsrtk/data/Network_checkpoints_segmentation/Unet.ckpt-20000\" -T \"0.5\" -o \"_masked.nii.gz\" \n",
+      "9lawi:\n",
       "... cmd: python brainExtractionTwoSteps.py -i \"/media/hkebiri/Storage/dti_t2_ANDRAS/export_normal/original-t2/NIFTI_HK/TestBrainExtraction/6_twoSteps/Image1_worldmatrixModified_lr.nii\" -c \"/home/hkebiri/Desktop/mialsuperresolutiontoolkit/pymialsrtk/data/Network_checkpoints_localization/Unet.ckpt-88000\" -t \"0.49\" -C \"/home/hkebiri/Desktop/mialsuperresolutiontoolkit/pymialsrtk/data/Network_checkpoints_segmentation/Unet.ckpt-20000\" -T \"0.5\" -o \"_masked.nii.gz\" \n",
-      "200819-18:26:52,511 nipype.workflow INFO:\n",
+      "200826-16:58:20,347 nipype.workflow INFO:\n",
       "\t [Node] Finished \"BrainExtractTwoStepsWorfklow.Multiple_Brain_extraction\".\n"
      ]
     }

notebooks/histogramNormalization.ipynb

@@ -295,7 +295,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.7.2"
+   "version": "3.7.3"
   }
  },
  "nbformat": 4,

pymialsrtk/interfaces/preprocess.py

@@ -661,16 +661,19 @@ class BrainExtraction(BaseInterface):
     input_spec = BrainExtractionInputSpec
     output_spec = BrainExtractionOutputSpec
 
-    def _run_interface(self, runtime): 
+    def _run_interface(self, runtime):
+        _, name, ext = split_filename(os.path.abspath(self.inputs.in_file))
+        out_file = os.path.join(os.getcwd().replace(self.inputs.bids_dir,'/fetaldata'), ''.join((name, self.inputs.out_postfix, ext)))
+
         try:
             self._extractBrain(self.inputs.in_file, self.inputs.in_ckpt_loc, self.inputs.threshold_loc,
-                             self.inputs.in_ckpt_seg, self.inputs.threshold_seg, self.inputs.out_postfix)
+                             self.inputs.in_ckpt_seg, self.inputs.threshold_seg, self.inputs.bids_dir, self.inputs.out_postfix)
         except Exception as e:
             print('Failed')
             print(e)    
         return runtime
 
-    def _extractBrain(self, dataPath, modelCkptLoc, thresholdLoc,modelCkptSeg,thresholdSeg, out_postfix):
+    def _extractBrain(self, dataPath, modelCkptLoc, thresholdLoc,modelCkptSeg,thresholdSeg, bidsDir, out_postfix):
 
         #Step1: Main part brain localization
         normalize = "local_max"
@@ -680,6 +683,7 @@ def _extractBrain(self, dataPath, modelCkptLoc, thresholdLoc,modelCkptSeg,thresh
         border_y = 15
         n_channels = 1
 
+
         img_nib = nibabel.load(os.path.join(dataPath))
         image_data = img_nib.get_data()
         images = np.zeros((image_data.shape[2], width, height, n_channels))
@@ -865,7 +869,6 @@ def _extractBrain(self, dataPath, modelCkptLoc, thresholdLoc,modelCkptSeg,thresh
 
 
         subImages = np.zeros((images.shape[0], width, height))
-        print(images.shape)
         for ii in range(images.shape[0]):
             subImages[ii, :, :] = cv2.resize(images[ii, x_beg:x_end, y_beg:y_end,:], dsize=(width, height))
         print(images.shape)
@@ -890,17 +893,20 @@ def _extractBrain(self, dataPath, modelCkptLoc, thresholdLoc,modelCkptSeg,thresh
                 pred3dFinal[idx, x_beg:x_end, y_beg:y_end,0] = pred_bin.astype('float64')
 
                 #pred3d.append(pred_bin[0, :, :, 0].astype('float64'))
-            pppp = False
+            pppp = True
             if pppp:
                 pred3dFinal = self._post_processing(np.asarray(pred3dFinal))
             pred3d = [cv2.resize(elem, dsize=(image_data.shape[1], image_data.shape[0]), interpolation=cv2.INTER_NEAREST) for elem in pred3dFinal]
             pred3d = np.asarray(pred3d)
             upsampled = np.swapaxes(np.swapaxes(pred3d,1,2),0,2) #if Orient module applied, no need for this line(?)
             up_mask = nibabel.Nifti1Image(upsampled,img_nib.affine)
-            nibabel.save(up_mask, dataPath.split('.')[0]+out_postfix)
+
+            _, name, ext = split_filename(os.path.abspath(dataPath))
+            save_file = os.path.join(os.getcwd().replace(bidsDir,'/fetaldata'), ''.join((name, out_postfix, ext)))
+            nibabel.save(up_mask, save_file)
 
     #Funnction returning largest connected component of an object
-    def _extractLargestCC(image):
+    def _extractLargestCC(self, image):
         nb_components, output, stats, centroids = cv2.connectedComponentsWithStats(image, connectivity=4)
         sizes = stats[:, -1]
         max_label = 1

setup.py

@@ -13,7 +13,8 @@
           "pymialsrtk.pipelines.anatomical"]
 
 package_data = {"pymialsrtk":
-                ['data/Network_checkpoints/*']
+                ['data/Network_checkpoints/Network_checkpoints_localization/*',
+                'data/Network_checkpoints/Network_checkpoints_segmentation/*']
                 }
 
 # package_data = {'cmtklib':

src/BrainLocalizationExtraction/brainExtractionTwoSteps.py

@@ -254,6 +254,7 @@ def extractBrain(dataPath, modelCkptLoc, thresholdLoc,modelCkptSeg,thresholdSeg,
         pred3d = np.asarray(pred3d)
         upsampled = np.swapaxes(np.swapaxes(pred3d,1,2),0,2) #if Orient module applied, no need for this line(?)
         up_mask = nibabel.Nifti1Image(upsampled,img_nib.affine)
+        print(dataPath.split('.')[0]+out_postfix)
         nibabel.save(up_mask, dataPath.split('.')[0]+out_postfix)
 
 #Funnction returning largest connected component of an object