Structural MRI (Magnetic resonance imaging)

October 25, 2009

tags: brain, hippocampus, MRI, quantitative, structural magnetic resonance imaging

3854685038_2ff4e74fe4

It was the August of 1997. I was doing my PhD at Department of Anatomy, Bursa, Turkey. It was a very hot day. I was reading some textbooks for my PhD dissertation topic. My brain was melting in my skull. It was like a beef steak in a pot burning. I decided to go the library where there is air-conditioner, and I would do some literature search as well. I do not know why I did that, but I entered the key words “anatomy” and “psychiatry”. Well, I had no idea what I was expecting to find. Wait a minute! There were some articles; particularly on schizophrenia. Here I was a happy camper at Anatomy Department teaching and learning anatomy, and there was my other passion psychiatry research which I had no idea on. It took me seconds to be back at the Department from the library. Since that day, for the last 12 years, I have been doing research on the anatomy of psychiatric disorders, particularly on mood disorders (i.e., major depression and bipolar disorder).

This is my story. But what is the story of “the anatomy of psychiatric disorders”? Here we go.

First of all, I can talk hours and hours on that, and I do promise to summarize everything in the following six paragraphs.

Wouldn’t it be nice to examine the brain of a person when the person is living? This question led into a big change in neuroscience. Before, it was an animal study, or a study during a brain operation, or post-mortem study (the brain of the deceased patient). But after the 80s with computerized tomography (CT) , and late 80s we see that brain researchers started to use magnetic resonance imaging (MRI); it was non-invasive and there was no radiation exposure; and most important of all a high quality of resolution compared to that of CT. As a result, we had the chance to see what was happening in the brains of patients with neurology and psychiatry, and neurosurgery while they were alive.

When I say structural MRI (magnetic resonance imaging; sMRI 1-2) I mean the quantitative measurements (mostly volume of any given brain structure) in the brain. We use this term very often, so I have left it as it is but you should know that it is about the brain. So you got the definition, right? Brain structures’ volumes (or very rarely areas ; particularly in the past just before 90s) are measured through MRI in a group of patients with a psychiatric (or neurological) disorder and at least age- and gender- matched healthy controls. We compare the volumes and see if there is any change; mostly decrease in that brain structure in the psychiatric patient group compared to a control group.

Structural MRI has the importance of detecting the affected structure in any psychiatric disorder. This was the first benefit from sMRI. For example, it is hippocampus for schizophrenia, it is amygdala for autism, etc.

The first question to be answered would be “is there any change in the size “volume” of that structure in the patient group. Second would be why. This comes to the famous question of quantitative MRI (another name for structural MRI) egg or chicken? So let’s say we are measuring the most famous structure to be measured in the brain “the hippocampus” on which I have spent 12 years. Hippocampus will be another post, but it is about memory and emotional regulation, so let’s say we are measuring it in depression (not surprising,no? considering its main functions). Hippocampal volume is decreased in patients with major depression compared to matched healthy controls. That is our finding.Ok? So are the patients depressed because they have smaller hippocampus or they have smaller hippocampus because of their depression. Here is a neat post from Dr. Kramer @ Psychology Today on that topic. You can ask this question in any psychiatric disorder.

Here in Canada, I have been working on a dataset where one of the main questions (maybe the main) was the effect of medication on these changes in the volume. Another question? Effect of genetics. Research is to ask the right questions after all. There are more questions you can ask through structural MRI, but it has been loosing its importance in the neuroimaging world since the beginning of 21st century.

Before I finish, I will answer the question: how are you measuring the volume of a brain structure on MRI? Let’s go back to history for that. Bonaventura Cavalieri was an Italian mathematician. He came up with the Cavalieri’s principle that set the rules of the scientific discipline; stereology. What is that? Simply it is used for measuring the volume of something from a two-dimensional environment; let’s say an image, a picture. I was at a stereology course in Samsun, Turkey in 1997. Let me give a simple example they gave us at the course. Take a cucumber. No, don’t eat it. We are going to measure it. Cut the cucumber in slices, with exactly the same slice thickness. Let’s say 1 cm. Now measure the area of each slice, and multiply it with your slice thickness, well you have an estimate volume of the cucumber. Of course, stereology is not all about this, but sMRI uses this principle of stereology. It is not difficult to explain what I am doing as a researcher when I am at a party. But, the difficult task is to define the brain structure of which I need to trace in each slice the structure is seen. When I say trace I mean drawing its boundaries with a specific program; our program was AFNI; a free program that runs in Linux. The program gives you the data you need when you are done tracing the structure so you get the volume. The boundary in the four directions (e.g., above, below) in a scan like this. Of course, your important task would be to decide in which slice I should start measuring and till which slice I should keep on measuring. Here is my volumetric measurement protocol of hippocampus on MRI.

Here is my very first structural MRI paper published in the first issue of Neuroanatomy.

Here is a more recent one titled ” A meta-analysis examining clinical predictors of
hippocampal volume in patients with major depressive disorder“

If you are a beginner, I recommend you to check the web pages of the big labs where they study the volumetric analysis of brain structures in psychiatric disorders or enter in PubMed; -the google of medical research- the keywords “MRI” “volume” “the brain structure you are interested in” and “the psychiatric disorder you are interested in”. And see what happens…

Below is a list of 19 big important players around the world who study structural MRI. The names are listed randomly.

Ludberg Tebartz van Elst

Freiburg, Germany

Epilepsy; mood disorders, particularly major depression

Chris Pantelis

Melbourne,Australia

Schizophrenia & mood disorders

Arthur Toga

Los Angeles, USA

Schizophrenia

Raquel E. Gur

Pennsylvania, USA