Friday, January 22, 2010


I was asked to explain some of the tests we use to evaluate the brain. This isn’t a topic easily covered in one installment, so it’ll be in segments.

The first attempt to look at the brain in a living human was the X-ray. But a standard front-to-back, side-to-side X-rays don’t give a picture of the brain at all. They show only the bones of the skull and the holes nerves pass through on their way to other structures (for example, the nerves that connect the eyes to the brain). It was the advent of small computers that advanced brain imaging.

For years before computers radiologists used the technique of tomograms for more detail than routine X-rays. A tomogram is a “slice” of a larger image. Imagine cutting a snapshot into slices. Arrange all those slices in order and you have a complete picture. Well, you can make slices of the head with X-rays too. Move an X-ray beam in an arc and there is one small point (focal point) through which all the beams pass. The further from that point, the more spread out (or out of focus) the beams are. Radiologists learned that this concept could be used to make very thin and focused X-ray “slices” of the brain. By moving an X-ray beam in an arc over the head the focal point will be in focus but the surrounding areas will be blurred. If you repeat this process at step-wise distances through the head, you would end up with “slices,” or images that are focued at varying distances through the skull. Tomograms of the skull were a huge advantage over regular skull films when looking for small changes in the bones caused by tumors.

Fast, reasonable sized, computers made it possible for these tomograms to be merged together to come up with 3D images. They also allowed X-ray to be taken and then displayed electronically without the use of film. As the speed of computers increased so did the number of slices that could be made in a reasonable period of time. More slices per picture resulted in improved detail. And as detail improved, so did the ability to show differences in brain matter instead of just bone detail. Because computers were used to merge tomograms, the technique is called computerized tomography, or CT scan. (Originally, the term computerized axial tomography, or CAT scan was used). Basically, CT scans are very sophisticated skull X-rays.

Magnetic Resonance Scans (MRI) make tomographic brain pictures from magnetic energy instead of X-rays. Remember the cartoon picture of an atom with electrons orbiting a nucleus. Because each electron is a negative charge, they try to stay away from each other like north poles of two magnets. If you place an atom in a very strong magnetic field the electrons are forced to align their orbits more than normally occurs. But the moment the magnetic field stops, the electrons fly back to their original orbits and in doing so, send off a small radio signal. Very sensitive radio receivers can pick up these signals. Because atoms from different elements (say, carbon and oxygen) have different numbers of electrons, the radio signals for different compounds are characteristic. Just like different radio stations transmit on their own frequency. Water has a different signal than protein or fat. These differences allow the MRI to give us very detailed pictures of the brain and allow us to see changes, such as tumors.

Although CT and MRI scans give detailed pictures of brain structure they tell us nothing about brain function. In contrast, PET Scans (positron emission tomography) provide us with information about brain activity, but without good structural detail of an MRI. PET scans are produced by injecting a chemical into the body called a radionucleotides, which are substances that emits activity that can be detected by Geiger-counter like sensors. Chemicals, such as glucose, can be made to emit positrons.

If a person does a task, for example move their right index finger, the brain region controlling that task works harder than other brain areas that remain idle, and the nerve cells producing the task require more glucose and oxygen. This change in glucose metabolism is what PET scans detect.

SPECT scans (single-photon emission computed tomography) are similar to PET scans but use less expensive equipment and a different technique. SPECT scans are primarily used to show blood flow through the brain and can be used, for example, to detect small strokes.

In summary, PET and SPECT scans give us information about blood flow and metabolism whereas CT and MRI show us brain structure. With very sophisticated computers, the pictures obtained from these different studies can be combined. For example, the information from an MRI and PET scan can be combined and the information used to help diagnose various brain problems.


  1. In your book DEAD HEAD, I was fascinated by how you laid out the challenges the surgeon, Russell Lawton, faced in keeping the injured man's severed head alive and functioning. You made it seem totally plausible, even down to devising a way for the head to communicate with his fellow terrorists. Ethical issues aside, could you ever see a time where a brain could be transferred to a host body, or perhaps just the thoughts and memories, not the actual brain?

  2. Hi Allen - Your post of January 22 was very helpful in describing the technical basis and practical uses of CT, MRI, Pet and SPECT. This explanation makes it clear what to expect and why when we visit our neurologists.
    Best regards - Vicky HS