Welcome to NI-edu¶
NI-edu is a collection of neuroimaging-related course materials developed at the University of Amsterdam. This website contains, at the moment, information about two courses, “fMRI-introduction” and “fMRI-pattern-analysis”, which are completely free and open-source. The courses contain lectures, extensive computer labs/tutorials, and seminars. The materials documented here only contain the computer labs (i.e., interactive Jupyter notebooks).
Below, the contents of the two courses are described in more detail.
The purpose of this course is to teach you the basic concepts and methodology of functional MRI (fMRI) research. By the end of the course you will be able to walk through the full empirical cycle (design, acquisition, analysis and report) of an fMRI experiment. With regard to the type of analysis, this course focuses on standard univariate analysis. Also, you’ll learn how to actually implement neuroimaging analyses, which is a fairly “technical” process, including working on Linux, programming, (relatively simple) mathematics, and statistics. So prepare to work with code, encounter mathematical formulas, and apply statistical models!
The purpose of this course is to learn how to implement “pattern analyses” for analysis of fMRI data. It discusses the two most prominent pattern analyses: machine-learning based ‘decoding’ and representational similarity analysis (RSA). At the end of the course, you will know when and why to use for the different types of analyses (‘univariate’ vs. decoding/RSA) and how to implement them. Note that it is meant as a follow-up course to fMRI-introduction.
The materials for this course are not completely ready yet. Stay tuned!
This course assumes that you have some experience with basic descriptive and inferential statistics (e.g., multiple linear regression) and Python programming. In case you have never programmed (in Python) before, fear not! As part of another course, we created several introductory tutorials about Python and the most important Python packages for data analysis (i.e., Pandas, Matplotlib, and Numpy). These tutorials are listed on https://lukas-snoek.com/introPy (under the PYTHON (WEEK 1) header).
Apart from the aforementioned prerequisites, this course assumes you are willing to work hard and put in a substantial amount of time. As we believe in learning by doing, the tutorials contain a lot of exercises. Doing these exercises are essential in achieving this course’s intended learning objectives.