Running on Windows

[Gresliebear]

Is the source code private?

I am trying to use on Windows 10 got error

basename: Output_training
tempFold: C:\Users\lesli\AppData\Local\Temp\tmpp4dzwfok\Output_training
stopping docker with command: docker stop greedy_elephant
Error response from daemon: No such container: greedy_elephant
stopping docker with command: docker rm greedy_elephant
Error: No such container: greedy_elephant
docker pull projectelephant/server
Using default tag: latest
latest: Pulling from projectelephant/server
Digest: sha256:60e5e0ece0f794a6e853e83ab40f99b9e898445fec6ace6b8f596c973ad2ec7c
Status: Image is up to date for projectelephant/server:latest
docker.io/projectelephant/server:latest
exam_import_folder: C:\Users\lesli\AppData\Local\Temp\tmpp4dzwfok
dicom_import_folder: C:\Users\lesli\AppData\Local\Temp\tmpp4dzwfok
exam_export_folder: E:\Datasets\BraTS challenge\Output
nifti_export_folder: E:\Datasets\BraTS challenge\Output
backend_scripts\win_docker.cmd 3 C:\Users\lesli\AppData\Local\Temp\tmpp4dzwfok E:\Datasets\BraTS challenge\Output C:\Users\lesli\AppData\Local\Temp\tmpp4dzwfok E:\Datasets\BraTS challenge\Output 0
E:\BraTS Challenge\vessel_env\src\brats-toolkit\brats_toolkit\util
starting docker!
Traceback (most recent call last):
  File "E:\BraTS Challenge\RedTinSaintBernard-for-BraTS2021-challenge\single_preprocessing.py", line 48, in <module>
    prep.single_preprocess(t1File=t1File,
  File "e:\brats challenge\vessel_env\src\brats-toolkit\brats_toolkit\preprocessor.py", line 106, in single_preprocess
    self.batch_preprocess(exam_import_folder=dockerFolder, exam_export_folder=dockerOutputFolder, mode=mode,
  File "e:\brats challenge\vessel_env\src\brats-toolkit\brats_toolkit\preprocessor.py", line 121, in batch_preprocess
    start_docker(exam_import_folder=exam_import_folder, exam_export_folder=exam_export_folder,
  File "e:\brats challenge\vessel_env\src\brats-toolkit\brats_toolkit\util\docker_functions.py", line 63, in start_docker
    subprocess.run(command, cwd=cwd)
  File "C:\Python39\lib\subprocess.py", line 505, in run
    with Popen(*popenargs, **kwargs) as process:
  File "C:\Python39\lib\subprocess.py", line 951, in __init__
    self._execute_child(args, executable, preexec_fn, close_fds,
  File "C:\Python39\lib\subprocess.py", line 1420, in _execute_child
    hp, ht, pid, tid = _winapi.CreateProcess(executable, args,
FileNotFoundError: [WinError 2] The system cannot find the file specified```

[neuronflow]

no source code is public, please feel welcomed to read the manual.

can you please specify what you are trying to do and how?

[Gresliebear]

ya, I found the link at the bottom, my bad. Anyways I am trying to do preprocessing on the 2020 Brats dataset because I want to try attempting the challenge for 2021 this year. I am running the Brats toolkit on windows 10 but I have read in all the data so I build a for loop to process it all at once.

import os
import pandas as pd
from brats_toolkit.preprocessor import Preprocessor
# instantiate
prep = Preprocessor()

## convert mapping info
## survial 
name_mapping = r"E:\Datasets\BraTS challenge\MICCAI_BraTS2020_TrainingData\name_mapping.csv"
survival_info = r"E:\Datasets\BraTS challenge\MICCAI_BraTS2020_TrainingData\survival_info.csv"

df_name_mapping = pd.read_csv(name_mapping)
df_survival_info = pd.read_csv(survival_info)

root_path_train = r"E:\Datasets\BraTS challenge\MICCAI_BraTS2020_TrainingData"
outputDir = r"E:\Datasets\BraTS challenge\Output\Output_training"
list_of_dir = os.listdir(root_path_train)

for name_of_file in list_of_dir:

    #if name_of_file contains .csv it skips iteration on the loop
    if name_of_file.endswith('.csv'):
        continue

    #We make new path tto list to for loop through we list that dir
    readable_path = os.path.join(root_path_train , name_of_file)
    list_of_zips = os.listdir(readable_path)

    # we for loop each folder
    list_sort = []
    outpath = os.path.join(outputDir, name_of_file)

    for zips in list_of_zips:

        readable_path_2nd = os.path.join(readable_path, zips)
        list_sort.append(readable_path_2nd)
        list_sort = sorted(list_sort)
        ## missing var for segmentation preprocessing # E:\Datasets\BraTS challenge\MICCAI_BraTS2020_TrainingData\BraTS20_Training_369\BraTS20_Training_369_seg.nii.gz 2 ??

    examName = name_of_file
    flaFile = list_sort[0] #  E:\Datasets\BraTS challenge\MICCAI_BraTS2020_TrainingData\BraTS20_Training_369\BraTS20_Training_369_flair.nii.gz1 flaFile
    t1File = list_sort[2] # E:\Datasets\BraTS challenge\MICCAI_BraTS2020_TrainingData\BraTS20_Training_369\BraTS20_Training_369_t1.nii.gz 3 t1File
    t1cFile = list_sort[3] # E:\Datasets\BraTS challenge\MICCAI_BraTS2020_TrainingData\BraTS20_Training_369\BraTS20_Training_369_t1ce.nii.gz 4 t1cFile
    t2File = list_sort[4] # E:\Datasets\BraTS challenge\MICCAI_BraTS2020_TrainingData\BraTS20_Training_369\BraTS20_Training_369_t2.nii.gz 5 t2File

    ## this code calls docker!
    ##dcm2niix conversion 
    prep.single_preprocess(t1File=t1File, 
        t1cFile=t1cFile, 
        t2File=t2File, 
        flaFile=flaFile,
        outputFolder=outputDir, 
        mode="cpu", 
        confirm=True, 
        skipUpdate=False, 
        gpuid='0')

so your preprocessing object from your tool kit is called at bottom to process each patient file basically. I notice it calls a docker object pull but I really don't know whats wrong with the code at this point I need to more reading into source code or check the variables I am passing to your object.

[neuronflow]

Hmm the Brats 2020 data is already preprocessed and skullstripped? There is no need to do it again? Can you elaborate please?

On Fri, 9 Jul 2021, 21:47 Leslie Wubbel, @.***> wrote:

ya, I found the link at the bottom, my bad. Anyways I am trying to do preprocessing on the 2020 Brats dataset because I want to try attempting the challenge for 2021 this year. I am running the Brats toolkit on windows 10 but I have read in all the data so I build a for loop to process it all at once.

import os import pandas as pd from brats_toolkit.preprocessor import Preprocessor

instantiate

prep = Preprocessor()

convert mapping info

survial

name_mapping = r"E:\Datasets\BraTS challenge\MICCAI_BraTS2020_TrainingData\name_mapping.csv" survival_info = r"E:\Datasets\BraTS challenge\MICCAI_BraTS2020_TrainingData\survival_info.csv"

df_name_mapping = pd.read_csv(name_mapping) df_survival_info = pd.read_csv(survival_info)

root_path_train = r"E:\Datasets\BraTS challenge\MICCAI_BraTS2020_TrainingData" outputDir = r"E:\Datasets\BraTS challenge\Output\Output_training" list_of_dir = os.listdir(root_path_train)

for name_of_file in list_of_dir:

#if name_of_file contains .csv it skips iteration on the loop
if name_of_file.endswith('.csv'):
    continue

#We make new path tto list to for loop through we list that dir
readable_path = os.path.join(root_path_train , name_of_file)
list_of_zips = os.listdir(readable_path)

# we for loop each folder
list_sort = []
outpath = os.path.join(outputDir, name_of_file)

for zips in list_of_zips:

    readable_path_2nd = os.path.join(readable_path, zips)
    list_sort.append(readable_path_2nd)
    list_sort = sorted(list_sort)
    ## missing var for segmentation preprocessing # E:\Datasets\BraTS challenge\MICCAI_BraTS2020_TrainingData\BraTS20_Training_369\BraTS20_Training_369_seg.nii.gz 2 ??

examName = name_of_file
flaFile = list_sort[0] #  E:\Datasets\BraTS challenge\MICCAI_BraTS2020_TrainingData\BraTS20_Training_369\BraTS20_Training_369_flair.nii.gz1 flaFile
t1File = list_sort[2] # E:\Datasets\BraTS challenge\MICCAI_BraTS2020_TrainingData\BraTS20_Training_369\BraTS20_Training_369_t1.nii.gz 3 t1File
t1cFile = list_sort[3] # E:\Datasets\BraTS challenge\MICCAI_BraTS2020_TrainingData\BraTS20_Training_369\BraTS20_Training_369_t1ce.nii.gz 4 t1cFile
t2File = list_sort[4] # E:\Datasets\BraTS challenge\MICCAI_BraTS2020_TrainingData\BraTS20_Training_369\BraTS20_Training_369_t2.nii.gz 5 t2File

## this code calls docker!
##dcm2niix conversion
prep.single_preprocess(t1File=t1File,
    t1cFile=t1cFile,
    t2File=t2File,
    flaFile=flaFile,
    outputFolder=outputDir,
    mode="cpu",
    confirm=True,
    skipUpdate=False,
    gpuid='0')

so your preprocessing object from your tool kit is called at bottom to process each patient file basically. I notice it calls a docker object pull but I really don't know whats wrong with the code at this point I need to more reading into source code or check the variables I am passing to your object.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/neuronflow/BraTS-Toolkit/issues/13#issuecomment-877418233, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABVY46SV5UYKVESLI3KF45TTW5G3NANCNFSM477LIDHQ .

[Gresliebear]

like preprocessing data for a model like a deep learning model, you need to preprocess your data sometimes its for pngs, nitfti files, but are you saying that this specific toolkit was used to create the dataset?

[neuronflow]

No the Brats 2020 data was mostly created by a pipeline from UPenn with a slightly different skullstripper but the outputs are almost identical. We might have a future update incorporating the UPenn pipeline. I encourage you to read the btk paper to develop a better understanding of the three btk modules.

On Fri, 9 Jul 2021, 21:53 Leslie Wubbel, @.***> wrote:

like preprocessing data for a model like a deep learning model, you need to preprocess your data sometimes its for pngs, nitfti files, but are you saying that this specific toolkit was used to create the dataset?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/neuronflow/BraTS-Toolkit/issues/13#issuecomment-877421474, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABVY46X725WLBSNVIB6UKFTTW5HUHANCNFSM477LIDHQ .

[Gresliebear]

ok so your saying that the dataset doesn't need any type of preprocessing? that the BraTS toolkit doesn't already provides? I will do more reading but btk modules? btk papers, I am sorry what resources are you talking about I haven't found these yet.

I will write my own preprocessing methods for our model but I am not trying to reinvent the wheel or deviate from common methods agreed upon by research, does that make sense?

[neuronflow]

https://github.com/neuronflow/BraTS-Toolkit#citation

Btk preprocessor can help you to preprocess Mr data so it can be segmented using BraTS algorithms.

On Fri, 9 Jul 2021, 22:02 Leslie Wubbel, @.***> wrote:

ok so your saying that the dataset doesn't need any type of preprocessing? that the BraTS toolkit doesn't already provides? I will do more reading but btk modules? btk papers, I am sorry what resources are you talking about I haven't found these yet.

I will write my own preprocessing methods for our model but I am not trying to reinvent the wheel or deviate from common methods agreed upon by research, does that make sense?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/neuronflow/BraTS-Toolkit/issues/13#issuecomment-877425735, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABVY46XJJA2QL6PNIDIVJR3TW5IVJANCNFSM477LIDHQ .

[Gresliebear]

ok I guess that makes sense? still confused I will continue trying anyways, Thank you for the help.