Brain Imaging Research: Structural MRI
The Autism Center of Excellence has a long history of conducting studies aimed at better understanding brain structure and development in individuals with autism. When autism was finally recognized as a biological disorder in the 1980s, ACE Director Eric Courchesne and his colleagues, were the first to show that the cerebellum is commonly observed to be abnormal in individuals with autism (see paper by Courchesne and colleagues, 1988).
Courchesne and colleagues were also the first to identify brain growth abnormality in the first year of life in infants with autism. Currently, researchers at the UCSD ACE are conducting MRI studies with toddlers between 12 and 36 months using the sleep fMRI method. These children are showing signs and symptoms of autism, but have not yet been definitively diagnosed with the disorder. This procedure is the only method currently available for directly examining brain development in children with autism right at the time when behavioral signs are starting to emerge.
Want to learn the major regions of the brain?
At the UCSD ACE we use state of the art brain imaging software to identify the major areas of the brain. The image below highlights some of the main subregions of the human cerebral cortex.
The ACE also studies typically developing children to further understand how the brain develops differently in autism. In the first study from our laboratory to utilize this method in 2001, (see Courchesne et al., 2001) we found that whole brain volumes appear to be normal at birth in children later diagnosed with autism, but that by 2 to 4 years of age, 90% of children later diagnosed with autism were found to have brain volumes larger than the normal average.
In addition, we found that the amount of white and gray matter in the cerebrum was significantly larger in children with autism compared to controls, whereas older children and adults with autism did not exhibit excessive volumes in these regions.
Later, in 2003, Courchesne and colleagues discovered that abnormal rates of brain growth in children with autism can start by age 1 or even younger. Understanding what causes abnormalities in the rate of brain growth in autism is one of our Center’s primary missions.
The 3-Phases of Brain Growth Model of Autism
Collectively, MRI and head circumference studies have given rise to a new hypothesis: autism involves three phases of early brain growth pathology.
- Early brain overgrowth at the beginning of life.
- Slowing or arrest of growth during early childhood.
- Although less established, we also believe that in a percentage of patients, degeneration may be present in some brain regions by preadolescence and continuing into adulthood.
This new theory of neural maldevelopment in autism highlights the first years of life as a key period when both malformation of neural circuitry is actively occurring and the first behavioral signs of autism are appearing. The second phase of brain development in autism shows heterogeneity, with the timing of arrested growth varying considerably across individuals such that a small number show only slight increases in brain size while others show macrocephaly (abnormally large head) and microcephaly (abnormally small head).
What defect drives this overgrowth remains unknown. The ACE proposes a tentative hypothesis that an excess of neurons (and their axonal and dendritic processes and synapses) in key regions that mediate higher-order social communication, emotion and language functions may drive overgrowth.
What part of the brain is enlarged in autism?
Other studies from the ACE have examined the cerebral lobes in the early life of autism (Carper et al., 2000; 2002). We observed that the early cerebral volume enlargement noted in 2 to 4 year-olds with autism appears to be greatest in the frontal lobes and closer to the normal average in the occipital lobes.
Furthermore, an inverse correlation has been observed between frontal lobe volume and size of the cerebellar vermis in children with autism between ages 3 to 9, meaning that children with an abnormally large frontal lobe volume also exhibited an abnormally small cerebellar vermis. These findings suggest that abnormal neural growth patterns early in development lead to a number of consequences in brain structure and function in autism.
What is MRI?
MRI stands for “magnetic resonance imaging” and is a technique used to take images of the human body using a strong magnet and radio waves. Although it can be used to take images of any part of the body, it is commonly used to image soft tissue organs, such as the brain, lungs, or liver.
MRI works by sending radio signals to a location of interest (e.g., brain) and receiving radio waves emitted back from that location within a large magnet. The properties of the received radio wave signals are different depending on the type of tissues being scanned.
These signals are later transformed into a single grayscale image by a computer, in which different features of the organ (e.g., gray matter and white matter in the brain) can be identified.
When placed in the presence of the strong magnetic field found in an MRI machine, the spin (or magnetic moment) of hydrogen atoms align themselves in the direction of the magnetic field. In short, they “line up.” To create brain images using MRI, short bursts of radio waves “flip” the hydrogen atoms over, knocking them out of alignment with the magnetic field.
Gradually, the hydrogen atoms line up again with the magnetic field, and as they do, they act as a small radio transmitter. Their signal is then picked up by the MRI machine.
Different tissues in the body take different amounts of time to realign. Dense tissue, for example, takes longer to realign than soft tissue.
To create a 3-D image as is commonly required for MRI studies of the brain, magnetic gradients of different strengths and different directions are applied on top of the main magnetic field and are turned on and off during the scan.
Thus, an MRI image is the result of a complicated interplay between radio frequency pulses and intermittently activated gradient fields, all of which are under computer control. Later, a mathematical technique called "Fourier transform" is used to translate the values obtained by the MRI machine into grayscale images.