Icon

Ksurfer_​workflow

The Java SSH Connector cannot share its connection with Python nodes because a live network socket cannot be serialized across the Java/Python process boundary — so we built a Python SSH Connector that stores credentials securely and lets each downstream node reconnect independently using the paramiko library.

1. ssh connector (python) node

after connection established we sent all the freesurfer commands to the HPC server where FreeSurfer installed.

  1. List remote files and folders (available in KNIME)

  2. Ksurfer subject validator (ensure the valid subjects) or selection of subjects

before commands execution, we need to initialise the remoted modules, activate license, python venv , FS env, etc.

  1. Ksurfer Feature Extractor ( running commands in FS env )

  2. Ksurfer hippocampus/Amygdala Reader (text files and structure them using python -> csv)

  3. Joiner (available in KNIME)

  4. Excel Writer/ Downstream ML (available in KNIME)





areavolumemeancurvthicknessthicknessstd
Cortical/Sub-cortical Measurements
cortical and sub-cortical measurements of the brain
Joiner
create a tmp file session save and read and write session IDduring connection write -> .jsonread the downstream nodesreset workflow delete the file.
SSH Connection (python)
Excel Writer
KSurfer Extractor
Top k Row Filter
check the subjects have valid freeSurfer measurements
Remote Subject Validator
Hippocampus Amygdala
Hippo/Amygdala Measurements
Row Filter

Nodes

Extensions

Links