View on GitHub

fmri-for-beginners

A course designed for anyone entering the field of fMRI-based cognitive neuroscience. It covers both the theory behind how we analyze the data and technical skills/knowledge to apply right away. This course is designed for those who are new to the field, but contains useful resources for all skill levels.

fMRI/The BOLD Signal

Now that you are all at least semi-pros on how MRI physics work, we can now go from structural to functional. Functional Magnetic Resonance Imaging(fMRI) uses the same imaging system as MRI, but we use different image acquisition parameters so that we can track changes in the brain over a short period of time.

fMRI works almost the same as structural MRI, but has two key differences:

  1. Higher temporal resolution (images are taken more frequently)
    • Instead of one high resolution image being taken over several minutes like a structural MRI, functional MRI takes several lower resolution MRI scans every couple of seconds
    • Typically in the range of 1.25-3.5 seconds (but can be outside of this) and depends on the image acquisition parameters
  2. There is lower spatial resolution (larger voxel sizes, worse SNR)
    • With a lower TR, typically comes lower spatial resolution or SNR
    • Lower spatial resolution isn’t “bad” spatial resolution, but it can be a limiting factor for some theories

So, how can we take a picture and tell what parts of the brain are active? Well for starters, you can’t do this with the naked eye. If you look at a montage of fMRI images(in the image below), you will likely see a heterogenous gray blob on your screen that looks like a mix between a Rorsharch Test and a brain. However, when we use complex computer programs, we are able to see what parts of the brain are active for certain tasks or even at rest.

Images of functional and anatomical MRI scans Image showing the difference between anatomical and functional MRI scans (Image from: https://www.newbi4fmri.com/tutorial-1-data)

Now, what exactly are these programs seeing for us? The Blood Oxygenation Level Dependent (BOLD) Signal of course! The BOLD signal shows changes in blood flow and blood volume in certain areas, which is used as an indirect indicator of neural activity.

But neurons use electrochemical signals not blood, right? Yes, that is correct. However, our neurons are energy powerhouses that use a lot of glucose and oxygen, so when a neuron goes from inactive to active, it needs some way to get more of these nutrients that it is quickly using. Luckily, our bodies are very smart, efficient machines. In order to provide the necessary amount of glucose and oxygen to those neurons, the brain redirects more bloodflow to areas those areas. The opposite goes for areas that are not active; they receieve less bloodflow.

This is the above information in simpler terms

While you may not find this video to be the most exhilarating, it is packed with useful information and very well done. This covers all of the basics in measuring the bold signal, HRF, and factors that can affect scans.

Principles of fMRI Part 1, Module 8: fMRI Signal & BOLD Physiology


This video really helps tie together some of the more practical aspects of MRI images that are relevant to know about:

fMRI Bootcamp Part 1 - Basics of MRI

Last Page - (What is an MRI) Next Page - (HRF)
Home