Python
is on its way to become the most used programming language in neuroscience
. It is easy to understand, can be learned rather quickly and has a very strong and helpful community behind it. There exist many amazing neuroimaging
software packages, such as Nipype
, Nilearn
, fMRIPrep
and more which facilitate the everyday life of a neuroscientist.
The goal of this 2-day workshop is to introduce participants to many different neuroimaging toolboxes
, the Python framework around it and everything you need to know about Nipype to start creating your own pipelines
and optimizing your workflows
.
# Save the workshop and nipype_tutorial content into your output folder
!cp -R /home/neuro/workshop/* /output/
Workshop overview10-11
Explore MRI data with Nibabel and NilearnHaving direct access to your neuroimaging data can be quite liberating. Nibabel
and Nilearn
allow exactly that. With those two neuroimaging packages, you can consider any brain image, a simple 3D/4D matrix of voxel values, which you can transform and move around like you would do with any other toolbox.
This section involves exercises and individual exploration of notebooks.
11-11:30
Introduction to NipypeIn this short introduction, we will show you what Nipype is and why you should use it. It contains a powerful short example that shows the strength behind Nipype.
11:30-12:30
Exploration of Nipype's building blocksNipype can be learned very quickly, but it's nonetheless important that you know about some of the main building blocks.
This section involves exercises and individual exploration of notebooks.
12:30-1:30
Creating a Nipype Pipeline from A-ZNipype preprocessing pipeline
Nipype analysis pipeline
This section involves exercises and individual exploration of notebooks.
1:30-2
Complex processing pipelines based on NipypeThere are many new and innovative neuroimaging
resources & softwares, such as BIDS
, fMRIPrep
, MRIQC
, OpenNeuro
, etc. And many of them wouldn't be possible without Nipype
and the open-source neuroimaging
community. In this section, we want to introduce you to some toolboxes and software packages that entail complex processing pipelines based on Nipype
. If you don't already use them yet yourself, you certainly want to be aware about them!
2-...
Open ended for questions10-11
PyBIDS and statistical analysis of fMRI dataThis section involves exercises and individual exploration of notebooks.
11-12
Functional connectivityIn this section we will focus on resting-state fMRI analysis
, in particular functional connectivity analysis
. The goal of this section is to show you some capabilities of Nilearn
.
This section involves exercises and individual exploration of notebooks.
12-1
Machine learning IWithin this section we will explore how some machine learning approaches
can be applied to fMRI
data. More precisely, we will have a look at "classic" machine learning
using Nilearn
and deep learning
using Keras
. The aim of this section is to provide you with some insights on how python
can be use to run these analyses on neuroimaging
data.
1-1:30
Machine learning IIBeside applying machine learning approaches
directly within neuroimaging
data more and more research also investigates the feasibility of using neuroimaging
data to predict
or discover
certain non-neuroimaging variables
such as different demographic information, disease & disorder status/progression, etc. . To give you a rough idea of how this can be done, we will go through an example within which we'll try to predict the age
of participants based on resting-state fMRI
.
This section involves exercises and individual exploration of notebooks.
1:30-2
The python-neuroimaging ecosystem This presentation more or less continues the presentation from the first session but focuses on the versatile set of python packages
available for neuroimaging
.
2-...
Open ended for questions