MAGNETIC RESONANCE IMAGING (MRI) HAS BECOME INDISPENSABLE for detecting tumors, torn ligaments and other anomalies under the skin. The technology has come a long way since its introduction in 1977, and the newest machines allow radiologists to peer deep into the body with near-cellular precision.  So researchers are focusing these powerful tools on the finer structures in the brain and shedding light on one of the body’s most puzzling and complex systems.

Anastasia Yendiki, assistant physicist at Massachussetts General Hospital, studies the role of white matter. This consists of bundles of axons—threadlike extensions of nerve cells—that weave together to form the brain’s underlying connections, sometimes called the connectome. That tangle is anything but straightforward, says Yendiki: “An image of it looks like a bowl of 500,000 spaghettis.”

She and her colleagues equipped their MRIs with high-gradient coils and other enhancements, allowing them to track the random motion of tiny water molecules through the brain. Their movements reveal white matter pathways, just as tracking the movement of cars can determine where roads are, says Yendiki. She has developed new software (christened TRACULA) to help her and other researchers sort out the course and strength of brain connections.

Scientists have already mapped the brain’s primary white matter highways, says Yendiki, “but we don’t know all the smaller roads leading on and off.” Yendiki is exploring these side roads—and how they vary in people with depression and other brain disorders. The Human Connectome Project, launched in 2009, is building a database of many connectomes that could help identify additional subtypes of complex diseases. For example, if the connectome is wired differently in people with depression, that information could be used to identify subtypes of the disease based on how they manifest in brain activity rather than in patterns of behavior, which is how psychologists currently identify them. This might guide more targeted and effective treatments for bipolar disorder, chronic depression and psychotic depression.

Other machines are equipped to take an even deeper dive. An MRI’s magnet causes molecules in the body to release energy that can then be recorded, decoded and finally translated into an image. A gold-standard clinical MRI uses a 3 Tesla magnet, which is about 300 times as strong as one you would put on the refrigerator door. But the newest machines, which weigh about 80,000 pounds and cost $7 million or more, use magnets capable of 7 Tesla and offer images with unparalleled resolution.

One of the 40 or so 7T machines is housed in the F.M. Kirby Research Center for Functional Brain Imaging, affiliated with Johns Hopkins University. Researcher Jun Hua is using it to study brain function in patients with schizophrenia, which affects about 3.5 million Americans.

He uses functional MRI (fMRI), a method that measures brain activity by looking at the flow of blood and oxygen. Previous fMRI studies showed that schizophrenia patients have an unusual level of activity between the thalamus—a neuron relay at the center of the brain—and parts of the brain’s surface. But it had been hard to pinpoint where in the thalamus those altered connections originated.

With the 7T MRI machine, Hua and his team were able to pinpoint the abnormal activity to specific sub-regions of the thalamus and trace it to regions on the brain’s surface that control sensory, motor and other functions. In addition to giving new insights about how schizophrenia manifests itself in the brain, such a tool might allow researchers to monitor a patient to see whether a particular therapy is helping to restore more normal brain patterns.

Hua also studies patients with Huntington’s disease, an inherited condition in which nerve cells gradually break down and impair cognitive and motor skills. Using 7T MRI, Hua discovered minute changes that could predict the onset of some common symptoms—such as impaired walking and involuntary jerking.

“We know the genetic mutation for this disease, so we can already test whether someone will develop it,” Hua says. “But we still don’t know when symptoms will appear.” Armed with Hua’s map, doctors could use fMRI to determine when to start treatment, before symptoms actually occur.

The brain is still a wilderness, and there may be generations before researchers fully grasp the intricacies of its architecture and workings. But in the next five to 10 years, says Hua, high-powered MRIs could help solve some of its more pressing mysteries.