fixed bugs regarding multiple connections, saving of metadata and loa…

…ding of metadata

Former-commit-id: 9caee64 [formerly 9caee64 [formerly ab236e5]] [formerly 9caee64 [formerly 9caee64 [formerly ab236e5]] [formerly bf97b7f [formerly bf97b7f [formerly 1528efc]]]]
Former-commit-id: bf97b7f
Former-commit-id: 82e0e186b9d508c398f798469326c643cd6527c2
Former-commit-id: 572070569f96d035e7a6f223bddc027a0517a0e0 [formerly a7c15fb]
Former-commit-id: 6fbeafe

imgtools/autopipeline.py

@@ -13,7 +13,7 @@
 import ast
 import datetime
 import numpy as np
-import json
+import pickle
 import shutil
 ###############################################################
 # Example usage:
@@ -84,7 +84,7 @@ def process_one_subject(self, subject_id):
            The ID of subject to process
         """
         #Check if the subject_id has already been processed
-        if os.path.exists(os.path.join(self.output_directory,".temp",f'temp_{subject_id}.json')):
+        if os.path.exists(os.path.join(self.output_directory,".temp",f'temp_{subject_id}.pkl')):
             print(f"{subject_id} already processed")
             return 
 
@@ -99,8 +99,8 @@ def process_one_subject(self, subject_id):
         for i, colname in enumerate(self.output_streams):
             modality = colname.split("_")[0]
 
-            #Taking modality pairs if it exists till _1
-            output_stream = ("_").join([item for item in colname.split("_") if item != "1"])
+            #Taking modality pairs if it exists till _{num}
+            output_stream = ("_").join([item for item in colname.split("_") if item.isnumeric()==False])
 
             #If there are multiple connections existing, multiple connections means two modalities connected to one modality. They end with _1
             mult_conn = colname.split("_")[-1].isnumeric()
@@ -109,7 +109,7 @@ def process_one_subject(self, subject_id):
             print(output_stream)
 
             if read_results[i] is None:
-                print("The subject id: {} has no {}".format(subject_id, ("_").join(colname.split("_")[1:])))
+                print("The subject id: {} has no {}".format(subject_id,colname))
                 pass
             elif modality == "CT":
                 image = read_results[i]
@@ -178,16 +178,16 @@ def process_one_subject(self, subject_id):
                 metadata[f"metadata_{colname}"] = [read_results[i].get_metadata()]
                 print(subject_id, " SAVED PET")
         #Saving all the metadata in multiple text files
-        with open(os.path.join(self.output_directory,".temp",f'temp_{subject_id}.json'),'w') as f:
-            json.dump(metadata,f)
+        with open(os.path.join(self.output_directory,".temp",f'temp_{subject_id}.pkl'),'wb') as f:
+            pickle.dump(metadata,f)
         return 
 
     def save_data(self):
-        files = glob.glob(os.path.join(self.output_directory,".temp","*.json"))
+        files = glob.glob(os.path.join(self.output_directory,".temp","*.pkl"))
         for file in files:
             subject_id = ("_").join(file.replace("/","_").replace(".","_").split("_")[-3:-1])
-            with open(file) as f:
-                metadata = json.load(f)
+            with open(file,"rb") as f:
+                metadata = pickle.load(f)
             self.output_df.loc[subject_id, list(metadata.keys())] = list(metadata.values())
         self.output_df.to_csv(self.output_df_path)
         shutil.rmtree(os.path.join(self.output_directory,".temp"))

imgtools/io/dataset.py

@@ -13,6 +13,7 @@
 import SimpleITK as sitk
 import warnings
 from imgtools.pipeline import Pipeline
+import datetime
 
 class Dataset(tio.SubjectsDataset):
     """
@@ -34,6 +35,7 @@ def load_from_nrrd(
             cls,
             path:str,
             transform: Optional[Callable] = None,
+            ignore_multi: bool = True,
             load_getitem: bool = True
             ) -> List[tio.Subject]:
         """
@@ -48,30 +50,43 @@ def load_from_nrrd(
         df_metadata = pd.read_csv(path_metadata,index_col=0)
         output_streams = [("_").join(cols.split("_")[1:]) for cols in df_metadata.columns if cols.split("_")[0]=="folder"]
         imp_metadata = [cols for cols in df_metadata.columns if cols.split("_")[0] in ("metadata")]
-        #Based on the file naming taxonomy
+        #Ignores multiple connection to single modality
+        if ignore_multi:
+            output_streams = [items for items in output_streams if items.split("_")[-1].isnumeric()==False]
+            imp_metadata = [items for items in imp_metadata if items.split("_")[-1].isnumeric()==False]
+        #Based on the file naming convention
         file_names = file_name_convention()
         subject_id_list = list(df_metadata.index)
         subjects = []
         for subject_id in tqdm(subject_id_list):
             temp = {}
             for col in output_streams:
-                extension = file_names[col]
                 mult_conn = col.split("_")[-1].isnumeric()
                 metadata_name = f"metadata_{col}"
                 if mult_conn:
                     extra = col.split("_")[-1]+"_"
+                    extension = file_names[("_").join(col.split("_")[:-1])]
                 else:
+                    extension = file_names[col]
                     extra = ""
                 path_mod = os.path.join(path,extension.split(".")[0],f"{subject_id}_{extra}{extension}.nrrd")
                 #All modalities except RTSTRUCT should be of type torchIO.ScalarImage
-                if col!="RTSTRUCT":
-                    temp[f"mod_{col}"] = tio.ScalarImage(path_mod)
+                if os.path.exists(path_mod):
+                    if col!="RTSTRUCT":
+                        temp[f"mod_{col}"] = tio.ScalarImage(path_mod)
+                    else:
+                        temp[f"mod_{col}"] = tio.LabelImage(path_mod)
                 else:
-                    temp[f"mod_{col}"] = tio.LabelImage(path_mod)
+                    temp[f"mod_{col}"] = None
                 #For including metadata
                 if metadata_name in imp_metadata:
                     #convert string to proper datatype
-                    temp[metadata_name] = df_metadata.loc[subject_id,metadata_name][0]
+                    meta = df_metadata.loc[subject_id,metadata_name]
+                    if pd.notna(meta):
+                        temp[metadata_name] = eval(meta)[0]
+                    else:
+                        #torch dataloader doesnt accept None type
+                        temp[metadata_name] = {}
             subjects.append(tio.Subject(temp))
         return cls(subjects,transform,load_getitem)
 
@@ -83,6 +98,7 @@ def load_directly(
             n_jobs: int = -1,
             spacing: Tuple = (1., 1., 0.),
             transform: Optional[Callable] = None,
+            ignore_multi: bool = True,
             load_getitem: bool = True
             ) -> List[tio.Subject]:
         """
@@ -94,6 +110,9 @@ def load_directly(
         input = ImageAutoInput(path, modalities, n_jobs)
         df_metadata = input.df_combined
         output_streams = input.output_streams
+        #Ignores multiple connection to single modality
+        if ignore_multi:
+            output_streams = [items for items in output_streams if items.split("_")[-1].isnumeric()==False]
         #Basic operations
         subject_id_list = list(df_metadata.index)
         # basic image processing ops
@@ -172,11 +191,11 @@ def process_one_subject(
 
 if __name__=="__main__":
     from torch.utils.data import DataLoader
-    # output_path = "/cluster/projects/radiomics/Temp/vishwesh/HN-CT_RTdose_test2"
-    input_path = "/cluster/home/ramanav/imgtools/examples/data_test"
+    output_path = "/cluster/projects/radiomics/Temp/vishwesh/HN-ctptdose_test2"
+    # input_path = "/cluster/home/ramanav/imgtools/examples/data_test"
     transform = tio.Compose([tio.Resize(256)])
-    # subjects_dataset = Dataset.load_from_nrrd(output_path,transform=transform)
-    subjects_dataset = Dataset.load_directly(input_path,modalities="CT,RTDOSE,PT",n_jobs=4,transform=transform)
+    subjects_dataset = Dataset.load_from_nrrd(output_path,transform=transform)
+    # subjects_dataset = Dataset.load_directly(input_path,modalities="CT,RTDOSE,PT",n_jobs=4,transform=transform)
     print(len(subjects_dataset))
     training_loader = DataLoader(subjects_dataset, batch_size=4)
     items = next(iter(training_loader))

imgtools/modules/pet.py

@@ -76,8 +76,8 @@ def get_metadata(self):
         except:
             pass
         try:
-            self.metadata["scan_time"] = datetime.datetime.strptime(self.df.AcquisitionTime, '%H%M%S.%f').__str__()
-            self.metadata["injection_time"] = datetime.datetime.strptime(self.df.RadiopharmaceuticalInformationSequence[0].RadiopharmaceuticalStartTime, '%H%M%S.%f').__str__()
+            self.metadata["scan_time"] = datetime.datetime.strptime(self.df.AcquisitionTime, '%H%M%S.%f')
+            self.metadata["injection_time"] = datetime.datetime.strptime(self.df.RadiopharmaceuticalInformationSequence[0].RadiopharmaceuticalStartTime, '%H%M%S.%f')
             self.metadata["half_life"] = float(self.df.RadiopharmaceuticalInformationSequence[0].RadionuclideHalfLife)
             self.metadata["injected_dose"] = float(self.df.RadiopharmaceuticalInformationSequence[0].RadionuclideTotalDose)
         except:

imgtools/ops/ops.py

@@ -276,8 +276,8 @@ def __init__(self,
 
         self.output = {}
         for colname in output_streams:
-            # Not considering colnames ending with _1
-            colname_process = ("_").join([items for items in colname.split("_") if items!="1"])
+            # Not considering colnames ending with alphanumeric
+            colname_process = ("_").join([item for item in colname.split("_") if item.isnumeric()==False])
             extension = self.file_name[colname_process]
             self.output[colname_process] = ImageFileOutput(os.path.join(root_directory,extension.split(".")[0]),
                                                            filename_format="{subject_id}_"+"{}.nrrd".format(extension))