A Bigger MRI Feels Less Claustrophobic

Prompt scans could render more patients eligible for clot-busting therapy

ScienceNews.org, November 3, 2010, by Nathan Seppa  —  MRI scans of stroke patients can indicate when the stroke occurred, a revelation that could allow more aggressive treatment to limit brain damage, French researchers report online November 2 in Radiology.

For a person arriving at a hospital with a stroke, the clock is ticking. When a clot obstructs an artery in the brain, millions of neurons are lost with each passing minute as tissue is starved of blood and oxygen. A clot-busting drug called tPA, or tissue plasminogen activator, can often dissolve the clot and free up the vessel. But the drug is generally considered safe to administer only in the first three to 4½ hours after a stroke begins (SN: 10/25/08, p. 16).

Stoke patients typically get a CT scan, which enables doctors to discern whether the stroke results from a blood clot or, less commonly, from a hemorrhage, which shows up as a dark mass on the CT, says neurologist Andrew Barreto of the University of Texas Medical School at Houston. MRI, or magnetic resonance imaging, is used much less often and usually only at large medical centers.

Unfortunately, a CT scan cannot pinpoint when a stroke began. Neither can many patients, either because they can’t recall exactly when their symptoms first appeared or because they woke up already in the throes of a stroke. In such cases, doctors “guesstimate” the stroke’s onset, Barreto says, but hesitate to give tPA if too many hours might have passed. Giving tPA too late won’t help tissue that’s already dead and risks causing a brain bleed. After the tPA window closes there is little doctors can do but monitor the patient.

MRI offers a more precise look into the brain of a stroke patient than a CT scan. In the new study, physicians Catherine Oppenheim and Mina Petkova of the University of Paris Descartes examined MRIs of 130 patients, average age 65, who had been admitted to Sainte-Anne Hospital in Paris from May 2006 to October 2008 with clot-based strokes that had documented onset times. About half had undergone an MRI within three hours of onset, while the others were imaged three to 12 hours after stroke symptoms started.

The doctors examined the MRIs without knowing when each was done. They applied three standard tests to what they observed, all measurements that assessed the extent of dead tissue in the brain resulting from a clot. One measurement, called fluid-attenuated inversion recovery, clearly distinguished between MRIs taken during the first three hours and those taken later. That measurement was accurate in about 90 percent of cases, whereas the other tests were less exact. “MRI could be used as a surrogate marker of stroke [duration] when the onset time of the stroke is unknown,” Oppenheim and her colleagues conclude.

“These data look provocative,” Barreto says. “If a CT scan shows no bleeding but subtle changes, you don’t always know what to do with the patient. That’s where an MRI is superior.” The MRI measurement that worked best in this study might reveal whether it’s advisable to give tPA six hours or more after onset, especially in people who awaken with a stroke, he says. Although an MRI can take 30 to 45 minutes to complete, Baretto says, that delay might be worthwhile if the readings expand the group of patients who could benefit from tPA. “It’s not perfect,” he says, “but it’s really good compared with other tests.”

The most detailed magnetic resonance images ever obtained of a mammalian brain G. Allan Johnson, Duke Center for In Vivo Microscopy

PopSci.com, GoogleNews.com, November 3, 2010, by Clay Dillow  —  Using an MRI system operating at six times the magnetic field of a conventional clinical scanner, researchers at the Duke Center for In Vivo Microscopy have gathered the most detailed magnetic resonance images ever captured of a mammalian brain. And while that may sound fairly wonky, the images – which will go into an online atlas open to researchers around the world – will allow researchers to slice and dice the brain digitally any way they want, helping researchers across disciplines run virtual experiments on the highly-detailed structure.

A typical MRI scan represents brain tissues in cubic units called voxels that can be thought of loosely as a 3-D pixel in a brain image. But since the new atlas images were taken at a resolution 300,000 times higher than those from a conventional MRI scanner, each voxel is much smaller and more detailed, shrinking from about 1-by-1-by-3 millimeters to about 20 microns per side (a micron being a millionth of a meter).

The images were taken of fixed brain tissue still in the craniums of eight different mice to minimize the distortions that can occur when the brain is removed or sliced for typical study. But using the tools in the Waxholm Space brain atlas – so named for the Swedish town where the idea for the atlas took shape – the brain can be digitally sliced at any angle, providing researchers with the ability to use the brain however they like without losing resolution.

Using three different data acquisition protocols while imaging, the team was also able to characterize 37 different brain structures, and by mashing up the data from the eight different mice they were also able to create a model for an average brain and a probabilistic brain. And that’s just the starting point for researchers; as new data is acquired it will be registered to the brain atlas site so other researchers can access it as well, fostering a system of information sharing that is far cleaner and more convenient than mailing mouse brains through the post.

Darn this aching back!”

Magnet.fsu.edu, by Kristen Coyne  —  That about sums up your thoughts as, flat on your back, clad in a thin hospital gown, you are guided into a dark, tunnel-like tube, a kind of medical solitary confinement where you’ll be spending the next 30 to 60 minutes. Your ears are stuffed with plugs, sound-cancelling headphones cradle your cranium, your extremities are stripped of watch, jewelry and other sentimental comforts, and you may feel a touch of claustrophobia. And it’s all because your orthopedist, worried about that chronic ache in your back, ordered an MRI scan.

Look on the bright side: MRI scans sure beat surgery, which was your only option 20-some years ago, before MRI became a routine, highly-valued diagnostic tool for doctors. In fact, when you take time to think about it, these machines are pretty amazing. And because you’ll be a captive audience while the technologist in the next room takes a bunch of fancy photos of your spine, we suggest you relax and let us tell you the story of the remarkable contraption in which you are, for the time being at least, stuck. We suggest you think of the cylinder as a kind of spacesuit, the machine as a spaceship, and you as an astronaut about to embark on a journey to the center of … your body!

You probably already have a vague idea of what’s going on. MRI – doesn’t that stand for Magnetic Resonance Imaging? Bull’s-eye. But what does that mean, exactly? Some type of X-ray, isn’t it?

Close, but not quite. MRI scanners, like X-rays and CT scanners, are basically machines doctors use to take pictures of your insides so that they can figure out what’s ailing you. But MRI doesn’t involve ionizing radiation, as do X-rays and CT scans. Rather, MRI takes advantage of something you have plenty of in your body: water. It is far more flexible than X-rays and CT scans, and can generate three dimensional images in any orientation and at any depth in the body.

PHYSICS FACTOID: In research and industry, MRI is known as NMR – nuclear magnetic resonance. It’s more or less the same process, but the medical establishment prefers the term MRI because some patients are scared off by the word nuclear.

The upshot is that MRI, for most applications, is far superior to other imaging tools in providing non-invasive images (and even chemical information) at high resolution. That’s why hospitals pay millions of dollars for the multi-ton behemoths, and spend hundreds of thousands more a year operating them. Since 1977, when the first MRI exam was conducted on a person, the procedure has become quite common. In 2003, some 60 million scans were performed using about 10,000 MRI scanners worldwide.

While X-rays remain useful for looking at bones, MRI scans are the diagnostic tool of choice for soft tissue – organs, ligaments, the circulatory system and (as you know) the spinal column and cord. They help physicians identify multiple sclerosis, tumors, tendonitis, strokes and many other conditions. What’s more, MRI technology is still in its infancy. Manufacturers are constantly improving scanner designs, and scientists are discovering new applications, from monitoring wine quality to detecting lies; one MRI study revealed that people used twice as many regions of the brain to tell lies as they did to tell the truth.

We know a lot about MRIs at the Mag Lab because we have the world’s strongest MRI scanner: Our 900 MHz nuclear magnetic resonance magnet produces a magnetic field of 21.1 tesla. As you’ll soon read, that’s far stronger than a hospital MRI scanner. We can’t fit humans in it, just laboratory animals such as mice and rats. But the MRI research done on those animals is helping scientists understand a wide variety of human disorders, from Alzheimer’s to cancer to muscle degeneration.

We have a lot of ground to cover in describing how this works. In the pages ahead, we’ll learn about the secret lives of hydrogen atoms, how radio waves can make you flip, and facts about superconductivity that will send shivers down your sore spine. It’s all fascinating, but try not to get too excited – you’ll blur your MRI scans!

Magnets with Muscle

Let’s start with a little tour of the metallic cylinder surrounding you (it’s called a bore, technically).

You’re in the center of a tremendous magnet, weighing tens of thousands of pounds and differing from the little magnets on your fridge in two fundamental ways. First, those fridge decorations are permanent magnets made of alloys. The MRI magnet surrounding you, on the other hand, is a superconducting magnet; it conducts electricity, thereby creating a magnetic field. Secondly, your fridge magnet has a fraction of the power of the one you’re in. Scientists measure magnetic strength in units called tesla and gauss – 1 tesla equals 10,000 gauss. The Earth’s magnetic pull is about .5 gauss. Your fridge magnet is about 10 gauss. The electromagnet you’re inside could be up to 3 tesla – 60,000 times the force of the Earth’s magnetic field.

PHYSICS FACTOID: The first MRI on a human was made in July 1977 by Dr. Raymond Damadian of New York.

Let’s take a moment to appreciate the “superconducting” part of that magnet, without which your MRI scanner would not be here. You could (and people do) make a permanent magnet with the strength to run an MRI. For the most part, however, these magnets are prohibitively huge and heavy. That leaves you the option of creating a magnet by running electrical current through wire coils – an electromagnet. The problem is the electrons making up that current are forever bumping into the fidgety atomic particles of the material through which they are traveling, slowing them down considerably. (Brush up with a quick review of electricity at this atomic level, if you need to). Given the resistance the current encounters, providing the vast amount of power required to overcome it and generate a magnetic field sufficient to operate an MRI would be prohibitively expensive.

This is where our hero, superconductivity, saves the day! Take special coils and surround them with something really, really cold – liquid helium, at 452.4 degrees below zero on the Fahrenheit scale, does quite nicely. The result? Those over-caffeinated atoms in the conducting wire are frozen into submission. Slowed to a virtual halt, they allow the current to sail right through the miles of wires snaking through an MRI scanner. This technology allows for the construction of hugely powerful magnets like the one surrounding you right now. Most clinical MRI scannerss use superconducting magnets. If you’re interested in learning more about this, you can read a more in-depth overview of superconductivity.

By the way, don’t let that little business about liquid helium worry you. It’s insulated in a vacuum, so you won’t need your parka. It wouldn’t be allowed, anyway; zippers, snaps, jewelry and other metals can become life-threatening projectiles in the vicinity of a magnet as powerful as this one. That’s why technicians are very careful to keep metals outside the exam room, and why people with pacemakers and aneurism clips can’t have MRI scans.

By now the technologist in the control room, who is talking you through the exam via an intercom, has started to place you into the main magnetic field in your scanner. The field is running horizontally through the bore from your head to your toes (or vice versa, depending on your position). Because your spine is what she’s interested in, you’re being positioned so that part of your back is in the middle (or isocenter) of the field.

You’d think that being in the middle of such a powerful force would make you feel different – tingly or something. It doesn’t. However, on an atomic level, it’s quite a different story, which takes us from the “M” of MRI to the “R” – Resonance. We’ll understand this better after first taking a close look at the molecules in your body.

Spin Control

You’re made up mostly of water, which means a large number of the atoms inside your body are hydrogen atoms. This turns out to be quite fortuitous, because hydrogen atoms happen to be built in such a way that they react dependably to the forces they will be subjected to inside this scanner. The first of these, as you by now know, is a main magnetic field. The second will be pulses of radio waves. But before we talk about those radio waves, let’s get better acquainted with the H of H2O.

In the nucleus of every hydrogen atom is a positively-charged proton that spins (or precesses, scientifically speaking) around an axis, much in the same way as a child’s top. This spinning generates its own tiny magnetic field, giving the proton its own north and south poles. Now, the nuclei of other atoms spin, too, but for a number of reasons (including, as we’ve mentioned, their sheer quantity), MRI is generally interested only in hydrogen atoms.

Under normal circumstances, these hydrogen protons spin about willy-nilly, on randomly oriented axes, as in the depiction below (more or less), showing hydrogen atoms before the MRI’s magnetic field is turned on.

– java –

However, when these atoms are placed in a more powerful magnetic field, it’s as if a drill sergeant blew a whistle: the protons line up at attention. Specifically, the axes realign with the more powerful magnetic field: Half of them face in the direction of the field, the other half in the opposite direction. In the Java applet above, click in the little box to turn the magnetic field on, and watch what happens. See the RF Pulse button? Don’t touch that dial! We’ll get to that in a minute.

PHYSICS FACTOID: In recent years, “open MRIs” have become increasingly available, offering more room for claustrophobic and large patients. Some radiologists caution the images may not be as good, as the machines often use weaker magnets.

Did we say half of the protons line up one way, and half the other? Well, not exactly. More precisely, a few more atoms (represented by the little blue guys) line up with the magnetic field in the low-energy configuration (north-south north-south) than in the north-north south-south configuration, which requires a bit more energy. (When we say “few,” we mean, in an MRI powered by a 1.5 tesla magnet, a measly 9 out of 2 million protons!) Those few “leftover” protons (the ones not cancelled out by a proton lined up in the opposite direction) are the ones your MRI scanner will be using. Think of these protons as the wallflowers left on the sidelines at a school dance after everyone else has paired up.

The Wonder of Waves

Now that the magnet has gotten the hydrogen protons lined up at attention, the scanner is ready to subject them to the next step, the one that will result in an actual signal.

You’ve probably been wondering about that coil the technologist placed under your back. No, that’s not one of the main magnetic coils (those are all inside the cylinder, beyond your view). Essentially, it’s a radio transceiver, also called an RF coil, which can communicate with your hydrogen atoms via radio frequency (RF) waves. These waves are close in frequency to those of your favorite FM station. In fact, the room in which the MRI scanner is located is probably shielded so that the local easy listening station doesn’t interfere with your images.

PHYSICS FACTOID: The vast benefits of MRI have not gone unnoticed by the folks in Oslo who dole out Nobel Prizes. In 2003, Paul C. Lauterbur and Peter Mansfield were jointly awarded the Nobel Prize in Physiology or Medicine for their discoveries related to the technology.

Your technologist is using that coil to send RF pulses at your spine. They are precisely timed (taking into account the type of tissue targeted and the fact that just the hydrogen atoms are of interest) to achieve the effect we’re about to describe. This, by the way, explains the “Resonance” part of MRI (told you we’d get to that!).

Remember those “unmatched” hydrogen protons – the ones (still depicted in blue, in the Java applet above) left hanging without a partner after the magnetic field caused them all to jump into alignment? Well, those protons absorb the energy of the RF pulses, which causes them to flip on their axes – still in line with the magnetic field, but now in the opposite direction, in the high-energy configuration. Scroll back to the applet and click the RF Pulse button to see how that works (make sure the magnetic field is still on). See how the unmatched protons flip as the RF pulses (denoted in red) are turned on?

It’s impressive enough that scientists figured out exactly how to make those hydrogen protons do that (the frequency needed is called the Larmor frequency). Now, the real magic happens: When the RF pulse stops, the protons release that absorbed energy, return to their previous alignments and, in so doing, emit a signal back to the coil (also in red, in the opposite direction). You may have noticed that has already happened in the applet. If you missed it, hit the “RF Pulse” button again for another look.

The signal gets turned into an electric current, which the scanner digitizes. The lower the water content in an area, the fewer hydrogen protons there will be emitting signals back to the RF coils. Different types of MRIs display this data differently, but in any case you get a variety of shades of grey that reflect the different densities. In some scans, the weaker the signal, the darker that part of the image will be. So bone will be fairly dark, while fat will be light.

Slicing and Dicing

What’s that? You need me to repeat that last part? I know – it can be hard to hear anything over the periodic hammering sound the scanner generates. Which reminds me – I’ve forgotten to mention anything about what’s making all that noise (and why you’re thankful for those earplugs).

PHYSICS FACTOID: MRI patients are sometimes injected with gadolinium, a contrast agent that can make abnormalities such as tumors clearer due to the element’s special magnetic properties.

Responsible for that racket are the gradient magnets. There are three of them in the scanner (called x, y and z), each oriented along a different plane of your body, all of them far less powerful than the main magnet. But what they lack in strength they make up for in precision. They modify the magnetic field at very particular points and work in conjunction with the RF pulses to produce the scanner’s picture by encoding the spatial distribution of the water protons in your body. When rapidly turned on and off (which causes that banging noise), the gradient magnets allow the scanner to image the body in slices – sort of like a loaf of bread. Using medical terminology, the transverse (or axial, or x-y) planes slice you from top to bottom; the coronal (x-z) plane slice you lengthwise from front to back; and the sagittal (y-z) planes slice you lengthwise from side to side. However, the x, y and z gradients can be used in combination to generate image slices that are in any direction, which is one of the great strengths of MRI as a diagnostic tool.

Starting to feel like chopped liver?

Let’s put a rush order on your MRI, so that you can see what we mean.

Shades of Gray

Here’s a picture (sagittal view) of your spine! Now it’s clear what the trouble is. See the dark disc that, unlike the others, protrudes into the spinal canal? That’s a herniated disc compressing the nerves of the spinal cord. Ouch!

Believe it or not, MRI scans can display more than 250 distinct shades of grey, each reflecting slight variations in tissue density or water content. It is in those subtle shades that radiologists unlock the secrets of the tissues. For example, abnormal tissue, such as a brain tumor, will look different than the normal tissue surrounding it. The technologist and radiologist have the ability to alter imaging parameters (like the timings of the RF pulse and gradients) to emphasize areas of injury or disease or to acquire higher image resolutions.

PHYSICS FACTOID: MRIs are most commonly used for cancer patients (about 35 percent of all scans) and patients with spinal problems (about 30 percent).

Now that your scan is over, you’re free to move around – do jumping jacks, dance a jig, celebrate your liberation. Unstop your ears, trade in the hospital gown for your street duds, reclaim your jewelry and watch. The MRI has not changed anything about your body or its chemistry. In fact, unlike X-rays or CT scans, you can have as many MRI scans as often as is necessary to diagnose your ailment and track your recovery after treatment.

We hope that the bad news about your back has been tempered by a newfound appreciation for the beauties of science and the wonders of technology. If we’ve whetted your appetite for more information on the topic, you might visit some of the Web sites listed below.

Links and Resources

Magnet Lab Resources

§                                 The World’s Strongest MRI Machine: The 900 MHz NMR Magnet
Scientists use animal models to study human health issues in this powerful machine.

§                                 More Than Skin Deep: MRI Research at the Mag Lab
When you have the most powerful MRI machine in the world, there’s a lot of exciting research going on. Read about some of the cutting-edge studies we do on neurodegenerative diseases, cancer, tobacco use, muscles and more.

§                                 What’s MRI
As defined by a Mag Lab expert (audio file).

§                                 Superconductivity: Current in a Cape and Thermal Tights
They don’t call it super for nothing. Once you get a superconductor going, it’ll keep on ticking like the Energizer Bunny, only a lot longer.

§                                 Magnets: From Mini to Mighty
If your knowledge of magnets ends with posting a to-do list on the fridge, add this to the list: Learn more about magnets! You can start here with a straightforward rundown of magnet types, uses and strengths, explained in a way that will help make the facts stick.

§                                 NMR Spectroscopy and Imaging at the Mag Lab
The Mag Lab’s NMR program supports state-of-the-art facilities and unique capabilities that are available to users pursuing research in solution and solid state NMR, MRI and in vivo magnetic resonance spectroscopy.

§                                 Advanced Magnetic Resonance Imaging and Spectroscopy Program
This Mag Lab user program, located at the Brain Institute of the University of Florida, contains facilities for the lab’s NMR program that complement the facilities at the lab’s headquarters in Tallahassee.

Other Web Resources

§                                 Magnetic Resonance Applications at High Magnetic Fields
The Mag Lab’s Sam Grant explains the topic in this online video from the Florida Education Channel

§                                 Imaging for Idiots (University of Manchester)

§                                 The Basics of MRI (by Joseph P. Hornak, Ph.D.)

§                                 What is Magnetic Resonance Imaging? (ehealthMD)

§                                 Magnetic Resonance Imaging (MRI) (radiologyinfo.org)

§                                 University of Florida Brain Institute

§                                 The Visible Human Viewery (University of Adelaide)

§                                 The MRI Game (Nobelprize.org)

§                                 MRI Glossary (European Magnetic Resonance Forum)

Thanks to the Magnet Academy’s scientific adviser on this article Sam Grant, an assistant professor of engineering with the Florida A&M University/Florida State University College of Engineering and an expert in MRI at the Magnet Lab.

Regions of your brain become active when you receive stimuli that could change your behavior.

SingularityHub.com, Fall 2010, by Aaron Saenz  —   Researchers at UCLA have discovered that MRI scans of your brain allow them to predict your future behavior better than you can. As discussed in the Journal of Neuroscience, people are generally about 50% accurate in predicting how they would adopt a new habit (using sunscreen in the experiment). By studying bloodflow to the precuneus and medial prefrontal cortex of the brain (both associated with self-reflection) scientists could accurately predict behaviour about 75% of the time. While very limited in its scope and sampling size, UCLA’s new insight into understanding how people will behave could have remarkable effects on everything from advertising to education. This is just one of several ways in which brain scans are poised to change our future.

We’ve seen a developing trend towards using MRI imaging to determine ‘what’s really going on’ in your mind. There have been ongoing efforts to use fMRI scans as lie-detectors for civil and criminal cases in the US, with little success so far. Researchers are attempting to be able to see ‘what the brain sees’ by directly reading activity in the visual cortex. Others are seeing how the technology may fit in with advanced security checkpoints. The UCLA research is unique however, because it does more than suppose a better understanding of what you’re thinking now, it suggests it might be able to predict what you are going to do.

The implications are huge. We could scan children as they are being taught and get a better understanding of what, if anything, they will retain later. The UCLA staff (in a University news article) reported that the technology may lead to ‘neural focus groups’ for marketing research. If advanced enough, the technology could even be used (perhaps erroneously) to help predict if criminals are likely to change their ways upon release. The concept of a machine being able to better understand your future behavior than you yourself could predict is a powerful and scary idea. The current reality, seen in the UCLA research, is actually rather limited however.

A variety of regions of interest in the brain show strong activity while someone is moving towards a behavioral change. This activity is a better predictor of change than people’s self-reported intentions.

In the study, volunteers in an MRI machine were shown public service announcements about the importance of wearing sunscreen. Their brains were monitored, especially in two regions of interest: the precuneus and medial prefrontal cortex. Those patients then predicted how often they would wear sunscreen in the next week. They were given sunscreen to make sure they would have access to it. Researchers followed up with a survey to track their actual use over the week. Only about half of volunteers accurately predicted their own future behavior. Meanwhile, scientists could use scans of the precuneus and medial prefrontal cortex of the brain to predict volunteer behaviour about three-fourths of the time.

How was this prediction made? The sample size of the study was small (10 men, 10 women) because of the costs and time associated with MRI technology. Rather than study the results of each volunteer and then predict an outcome, researchers waited until the results were in. They then randomly selected 10 people and used their scans and results to build a computer model for prediction which they then used on the remaining 10 people. Of course, they didn’t want to be limited to just one random selection, so they repeated this process for every possible combination of 10 models and 10 test subjects. That’s 184,756 repetitions if anyone’s counting. Taken as a whole, the MRI-based models ended up around 75% accurate in predicting the results of the other group from their scans.

This is actually a fairly reliable and interesting approach to the problem, but we shouldn’t be blind to its limitations. 20 people is a tiny sample set, no matter what statistical expertise you apply to it. The study also relied on volunteers to self-report their own sunscreen use during the test week. This introduces bias into the results, but presumably in favor of raising a volunteers own self-prediction accuracy (one assumes people will tend to lie to make themselves have been right in the past). The researchers acknowledge these limitations in their paper, and hope to pursue a larger testing group with better controls in the future.

The bigger limitation here, however, is that this early study into the phenomenon does not truly simulate a real-world experience. Being inside a two-ton machine while being shown slides on sunscreen isn’t the way that most of us are exposed to influencing media. It also doesn’t take into account long standing attitudes and histories with the product or idea being presented (sunscreen use). Even if we eventually get very good at building MRI models to predict people’s response to stimuli in a lab setting, that doesn’t necessarily mean we’ll be able to predict things in the more chaotic and dynamic outside world.

That being said, I’m still rather excited by this work. As mysterious as we may find our mind and consciousness, experiments like this one demonstrate that our brain is a fairly understandable organ. When regions of the brain are used heavily, you can reasonably predict that their specialties are affecting what we’re thinking. For example, in the UCLA study researchers noticed increase activity in the parts of the brain that handled self-reflection (like the precuneus and medial prefrontal cortex) predicted a change in behavior (using more sunscreen). The research team also found correlations between the regions of the brain that handle perspective of others, memory encoding, attention, visual imagery, motor execution and imitation, and affective experience with increased behavior change. Hinting that each of these is a factor in determining how and why we ultimately choose to do something new.

This is really fascinating stuff, and it shows that not only can we learn about the brain, but that we have a lot more to learn before we fully understand it. The predictive models at UCLA need years of more work before they are likely to find applications, but the potential that they could succeed is certainly there. Along with other uses for truth detection, image processing, and emotional state, the predictive models show that MRI may become one of the more versatile and powerful diagnostic tools of the future. Now if only someone could make it smaller and portable – the giant metal cocoon look is so last century.

[image credits: Falk et al, JoN 2010]
[source: *Falk et al, JoN 2010; UCLA News]
*Full PDF of the report is available via the Social Cognitive Neuroscience Lab at UCLA.

What is Functional Magnetic Resonance Imaging or FMRI?

Functional Magnetic Resonance Imaging or FMRI is a non-invasive technique for imaging the activation of brain areas by different types of physical sensation (sight, sound, touch, taste, smell) or activity such as problem solving and/or movement (limited by the machine). Thus, FMRI scans are an increasingly common tool for “brain mapping” in cognitive science.

How Are MRI Machines Constructed?

The construction of MRI Machines has evolved somewhat over the years. Two factors have influenced the development of MRI design; (1) the desire to enhance image quality and (2) The desire to make the scanners less confining for the patient/subject. Below are some of the design innovations introduced to make MRI scanners less claustrophobic and allow the subject more freedom for performing tasks while being scanned. Figure 1 shows a typical MRI scanner where the subject is nearly enclosed within the long tube of the scanner. Figure 2 shows how designers the shortened the scanner tunnel. Short-bored designs are less claustrophobic for the patient. These designs ease the sense of confinement, but still limit the subject’s ability to engage in tasks. Stand-up scanners like the one shown in figure 3 are more convenient for patients and allow imaging in the normal attitude and weight-bearing conditions. The scanner is lowered around the subject, who sits on an adjustable seat. Finally, figure 4 shows open MRI scanners, which allow for a greater range of subject tasks as well as easing the subject’s sense of confinement.

Though specifics of design vary, the basic elements of an MRI scanner remain pretty much the same (See below). The scanner consists of a large magnet (blue) that creates the primary magnetic field. Magnet strength in MRI systems is measured in units of magnetic flux density called a “tesla”. A telsa is enough magnetic force to induce 1 volt of electricity in a single-coil circuit during 1 second of time for every square meter. 1 tesla is equivalent to 10,000 gauss, another meaure of magnetic force defined as one line of force per square centimeter per second of time. Current magnet strength varies from 0.5-tesla to 2.0-tesla. However, researchers developed 3-tesla MRI scanners in the late 90’s which are becoming more common. To put those numbers in perspective, the Earth’s magnetic field is about 0.5-guass or .000005-tesla. In addition to the main magnet, there are also gradient coils (red). These gradient coils are electro-magnetic coils which technicians use to alter the main magnetic field at very precise points and for very precisely controlled times. Gradient coils can be changed so as to adjust the machine for the type of body material to be imaged. Finally, MRI scanners also incorporate radio frequency coils which can send a focused radio frequency pulse into the scanner chamber. Technicians can change the radio frequency coils to adjust for materials and body part.

What’s the Difference Between MRI and FMRI?

FMRI scans use the same basic principles of atomic physics as MRI scans, but MRI scans image anatomical structure whereas FMRI image metabolic function. Thus, the images generated by MRI scans are like three dimensional pictures of anatomic structure. The images generated by FMRI scans are images of metabolic activity within these anatomic structures.

MRI image of head (single slice) Temporal Sequence of FMRI scans (single slice) Three Dimensional Image of Brain Activation

What is Magnetic Resonance?

In order to understand how FMRI scans work one needs a rough understanding of the basic physical principles upon which the technology is built. The relevant physical principles are those involving the atom. Atoms are the smallest particles of an element which still possess the properties of the elements. For instance, helium is an element. The smallest bit of helium that still has the properties of helium is a helium atom. Atoms are very small. The diameter of an atom ranges from about 0.1 to 0.5 nanometers (0.0000000001 of a meter to 0.0000000005 of a meter). To put this in some context, if we think of the diameter of a single atom as the length of a meter stick, the corresponding length of the meter stick would be 10 billon meter sticks (approximately 14 round trips to the moon). Despite their diminutive dimensions, atoms are mostly empty space. The nucleus (center) of an atom has a diameter approximately 1/10,000 that of the diameter of the atom itself.

Most atoms are composed of three particles distinguished by their electrical charge: protons (positive), electrons (negative), and neutrons (neutral). Electromagnetic forces bind protons and neutrons together in an atom to form its center, i.e, its nucleus. The number of protons in an atom’s nucleus determines the atom’s elemental categorization. Hydrogen has the fewest protons with only one. Uranium has 92 protons. The number of neutrons is usually approximately equal to the number of protons, but there is variation in the number of possible neutrons in an atomic nucleus. Electrons circle around the nucleus. Since protons have a positive charge and electrons have a negative charge these particles attract each other, thereby creating the stable, electrically neutral structure of the average atom.

Not-to-scale model of a helium atom.



The electromagnetic forces that keep atomic structure relatively stable by keeping the electrons circling the nucleus also cause the nucleus to wobble or spin. That is, the nucleus of the atom spins around as in the above animation. Nuclear spin, or more precisely, the manipulation of nuclear spin is the basis for MRI imaging. If you follow astronomy, nuclear spin is similar to the wobble of distant stars used to infer the number, size, etc. of bodies orbiting the star.

If one places an atom within a magnetic field plane, i.e, subject it to magnetic forces along two of the three dimensions, then the nucleus will orbit around the third (vertical) axis. This is called precession and is depicted in the animation below:

Not-to-scale model of a helium atom.

When one causes nuclei to precess their spin will cause them to align themselves with the magnetic field. The spin of a nucleus is just like the ends of a bar magnet in that it can have a positive or negative value. Two negative or two positive ends of a magnet repel one another, but negative and positive ends attract each other. Similarly, all the negative spin atoms align themselves downward on the Z axis (towards the feet of the subject), and all the positive atoms align upward on the Z axis (towards the subject’s head). Each atom with a positive spin cancels out (renders undetectable) an atom with a negative spin. There remain, however, a few atoms that do not cancel one another out. At room temperature, there are always more positive spin atoms than negative spin atoms. These unmatched atoms are the important ones for MRI and FMRI.

Positive spin atoms are in a low energy state. The atoms achieve and equilibrium magnetization value along the direction of the magnetic field, i.e., the Z axis. By introducing a pulse of magnetic energy perpendicular to the main magnetic field in the form of a radio frequency pulse that is specific to the type of atom (usually hydrogen), the MRI machine causes the unmatched atoms to resonate. Resonating atoms absorb the radio energy as a photon and go to the higher energy state, i.e., they become negative spin atoms and the equilibrium magnetization value for the Z axis goes to 0. When the pulse is stopped, these atoms release their photon energy and “relax” back into the lower energy positive spin state. The signal that the MRI machine detects is the photon energy emitted by these unmatched atoms as they make a transition from the higher energy state to the lower energy state after the radio frequency pulse. The amount of time it takes for for the atoms to return to their equilibrium value is called the “spin lattice relaxation time” or (T1). T1 is, thus, a measure of the half-life of inverted spins.

If the technician uses the gradient magnets inside the MRI to alter the local net magnetization so that it is in the XY plane (cutting a very thin slice across the patient), the local net magetization rotates the Z axis (takes on positive and negative X and Y values) at a frequency called the Larmor frequency. The Lamor frequency equals the frequency of the photon which would cause a transition between the two energy levels of the nucleic spin. By again introducing a pulse of magnetic energy in the form of a radio frequency pulse that is specific to the type of atom, the MRI machine causes the unmatched atoms to resonate. The resonating atoms absorb the radio energy and go to the higher energy state, i.e., they become negative spin atoms relative to the XY axis (the transverse axis). The amount of time it takes for for the atoms to return to their equilibrium magnetization value along XY axis (transverse axis) is called the “spin-spin relaxation time” or T2. T2 is, as a result, measures the rate of change of spin phases. Whereas a typical T1 (spin lattice relaxation time) is approximately 1 second, the T2 (spin-spin relaxation time) is usually less than 100ms. This difference in the relative times is what makes T2 better suited than T1 for functional metabolic imaging.

Spin-spin relation time T2

Particularly important for FMRI is the measure of decay of transverse magnetization, T2* which takes into account two important factors: (1) molecular interactions and inhomogeneities in the magnetic field. FMRI creates the images or brain maps of brain functioning by setting up and utilizing an advanced MRI scanner in such a way that increased blood flow to the activated areas of the brain shows up on the MRI scans. The MRI scanners do not actually detect blood flow or other metabolic processes. Rather, blood flow alterations and/or associated metabolic processes in brain areas are indirectly inferred from the signal intensity contrast for a given brain region relative to both normal levels and levels immediately adjacent to the area in question. The intensity of an MRI signal is determined by the level of magnetic resonance, specifically what is called the BOLD effect (blood oxygenation level dependent) on T2*. Here’s how:

Magnetic fields are altered by the presence of any substance to some extent.

Many materials exhibit pronounced polarization in a magnetic field. The degree of this effect is referred to as the “magnetic moment” or “magnetic susceptibility”. Spatial and temporal variation in local concentrations of deoxygenated hemoglobin (blood cells not carrying oxygen or Hb) to oxygenated hemoglobin (blood cells carrying oxygen or HbO2) result in corresponding changes in magnetic susceptibility, which in turn cause the local T2* values to fluctuate. Oxygenated hemoglobin are diamagnetic (i.e., tend to take a position at right angles to the lines of magnetic force, and are repelled by either pole of the magnet), while deoxygenated hemoglobin is paramagnetic (i.e., takes a position parallel and proportional to the intensity of the magnetizing field). Thus, MRI is able to detect a small difference (a signal of the order of 3%) between the two types of hemoglobin. This is called a blood-oxygen level dependent, or “BOLD” signal. Researchers are currently exploring the precise relationship between neural activity and the BOLD signal. However, the basic story goes as follows:

Blood is delivered to the brain by arties and transported from the brain by veins. Not only is the actual blood volume relatively low in the brain, but the majority of blood volume is in the capillary bed–the very small vessels that connect arteries and veins. Capillaries are often so small that hemoglobin (blood cells) travel in single file. Whereas arterial blood has a high concentration of oxygenated hemoglobin as the blood cells pass through the capillary bed, the concentration of deoxygenated hemoglobin increases relative to oxygenated hemoglobin. Thus, as indicated in the diagram below, a gradient of highly oxygenated hemoglobin to highly deoxygenated hemoglobin runs across the capillary bed from arteriole to venule. As a result, a corresponding gradient in T2* values from longer T2* (diamagnetic HbO2-rich) to shorter T2* values (paramagnetic Hb-rich).

Capillary bed

Image modified from: http://www.bmb.psu.edu/courses/bisci004a/cardio/capbed.jpg

The relevant spatial unit for measuring local T2* for fMRI contrast called a “voxel“. Voxel comes from the contraction for volume element. A voxel is smallest unit of MRI reconstruction, and corresponds to a single pixel in an MRI display image. The relative ratio of deoxygenated to oxygenated hemoglobin within a voxel determines the T2* value for that voxel. Increases in metabolic function in a given brain region can trigger vasodilation (expansion of the vessel), thereby increasing oxygenated blood flow and altering the gradient of highly oxygenated hemoglobin to highly deoxygenated hemoglobin within the capillary bed. If one assumes that neuronal activation causes local vasodilatation absent a corresponding significant increase in oxygen metabolism, then increased oxygenated blood flow to a brain region results in a corresponding increase in local, intravoxel T2*. This T2* increase, then, causes a corresponding increase in image intensity.

How Researchers Use MRI Scanners to Expolit the BOLD Effect

The subject in a typical experiment lies or sits in the magnet and the researcher sets up a particular form of stimulation or task. For example, the subject may wear special glasses so that pictures can be shown during the experiment. Then, MRI images of the subject’s brain are taken. Like PET scans, each of these images are of a single slice of the brain which a computer combines into a 3-D image. On the first pass, researchers take a high resolution scan to be used as the backdrop against which the activated areas will be better differentiated anatomically. Next, the subject begins that stimulation or task and and series of the lower resolution BOLD scans are taken over time. BOLD scans are usually repeated every 2-5 seconds. The voxels in FMRI are approximately 2.5 mm each side.

Slices taken in succession

Series of slice images of brain

Image Courtesy of Wikipedia

Upon completion of the experiment, the computer takes the slice images from the MRI and uses mathematical transformations and reconstruction algorithms to render the images and then correct for distortion, subject movement, etc.. The computer compares and accurately maps the the low resolution images taken during the experiment to the high resolution backdrop. The final image illustrates areas of activity during the experiment as colored patterns overlaid upon the original high resolution scan. Researchers can also render the combined activation slice images into a 3-D images which can be rotated to any angle. For a really powerful example of how such imaging works look at the mpeg video made from images of children’s brains as they matured: UCLA Brain Maturation The movie is made from time-lapse imaging constructed from MRI scans of healthy children from 5 to their 20s. Red indicates more gray matter, blue less gray matter.

The Subtraction Technique for using FMRI for Complex Tasks

In many cases the tasks that cognitive scientists which to study–the target or primary task–also require that the brain perform other tasks–the secondary tasks. For example, the reasoning tasks we looked at during the logic lectures involved not only deductive or inductive reasoning tasks (primary or target), but also language comprehension tasks as well (secondary). In order to better isolate the areas of the brain involved in one task when the brain may have to perform several tasks, scientists perform FMRI scans on subjects during the performance of two tasks different tasks. The first task will represent the other tasks hypothsized to be operant (the secondary tasks) with the exception of the task of primary interest (target). For instance, in the reasoning studies done by Osherson et al. (1998) subjects performed both reasoning tasks and meaning tasks. Since language comprehension was clearly involved in evaluating the arguments, subjects were asked to peform the meaning tasks. As Osherson et al. express it, “The meaning task required subjects to examine premises and conclusion individually and determine whether any had anomalous content; it served as a baseline condition since no more than sentence comprehension was involved.” (1998, p. 370) The second task is the target task–the task of primary interest including the necessary secondary tasks. In the case of Osherson et al. subjects evaluated whether arguments were valid or invalid in the deductive logic task. Scientists then subtract the activation patterns in the performance of the secondary only task from the activation patterns during the target task (combined primary and secondary task). The remaining activation is then hypothesized to result exclusively from the primary or target task and not the secondary tasks.

Last Part of Samsung 3D Show on Side of Building

How do you sell 3D TVs? By showing them burst out of buildings, that’s how!

SingularityHub.com, November 3, 2010, by Aaron Saenz  –  Sometimes an idea is so good, that everyone has it at the same time. Sony, Samsung, and Toshiba, three major forces behind 3D television sets, each decided to promote their new TV technology with massive ‘3D’ projectors† that can display huge video demonstrations on the sides of buildings. Not only did they all use the same ‘3D’ projector marketing concept, but they all hired the same company to create it. Seriously. NuFormer, a multimedia firm in the Netherlands, helped Samsung promote their 3D TV in Amsterdam the same weekend in May it was putting on a show for Sony in Madrid. In September NuFormer helped Toshiba wow audiences at the IFA Fair in Berlin with another ‘3D’ projection show. I would mock these companies for their lack of originality except for one thing: the presentations totally rocked. Watch all three in the videos below. They may all use the same technology, but each one is its own eye-popping piece of art.

Samgsung hired NuFormer to destroy the historic Beurs van Berlage building in Amsterdam. The projectors cast images on the front of the edifice that made it look like it was crumbling, and slowly being replaced by nature. The demonstration was only up for two hours every night from May 20th to 22nd. As you can see in the video, the crowd was quite pleased with the results. Samsung took their ‘destruction by nature’ concept one step further and brought it online. A special channel on YouTube (Samsung3Devent) destroys itself in the same manner as the Beurs van Berlage, tearing the webpage down and replacing it with a butterfly catching game. The winner reportedly received a free Samsung 3D TV, but unfortunately the contest is now over. Oh well, at least the presentation is still available to watch:

Samsung used their YouTube channel to let those at home experience the fun of ‘3D’ marketing. Actually…this was pretty cool of them.

The really crazy thing is that the last night that the Beurs van Berlage was being bombarded by butterflies, football fans in Madrid were watching a ‘3D’ homage to the game they loved courtesy of Sony and NuFormer. “Imagine Football in 3D” was projected on to two different buildings at the Plaza Santa Anna and the Colegio San Augustin. 3D sport telecasts look to be one of the big selling points for 3D television sets, and Sony is smart to try to hook that branch of their target market early on. Watch Sony stimulate Spanish soccer fans in the video below:

IFA Fair is one of the largest consumer electronics shows in the world, and everyone brings their ‘A’ game to Berlin. Drawing a crowd and promoting Toshiba’s 3D home entertainment systems, NuFormer used their projection systems to transform two enormous boxes into a family den where the entertainment is leaping out of the TV. Not quite a building-sized project, but still very cool:

NuFormer isn’t the only company that specializes in these mammoth presentations (though apparently Sony, Samsung, and Toshiba all thought it was). We’ve even seen the pinball motif displayed in the Madrid presentation before. I’ve been really into building sized projection systems since we first covered them last year, and I think that the cool ‘3D’ video art from LCI, Urbanscreen, and individuals like Pablo Valbuena are as high quality as what you’ve seen above. This technology is all the more amazing because so many different groups around the world are working with it. Building sized projectors aren’t quite commonplace yet, but work like this shows that it could be soon. It’s going to be pretty intense when every third house or office complex has a full sized night time advertisement and art show. Actually I can’t wait, sounds awesome.

And as the for the other technology, 3D TVs? Well those are already here in force. You can expect Samsung, Sony, and Toshiba to be hyping these screens like crazy as we enter into the holiday season. Hopefully that will mean more cool building-sized ‘3D’ art projects. Not that such marketing gimmicks would actually convince me to purchase one of these first generation 3D devices. As I’ve said before, I don’t think they’re ready yet. I’m not buying a 3D TV. …I’m just not going to do it. …It would be a total waste of money…right?

†Note: I’m using ‘3D’ here to mean a flat image projected onto a surface in a way that makes it appear to have depth. This sort of perspective based ‘3D’ has been around for a long time. I’ll only refer to things as 3D (no single quotes) when they actually use a stereoscopic or other three dimensional effect that requires you to have two eyes to see.

[screen capture credit: Samsung]

[sources: NuFormer]