Silent, motion-resistant fMRI expands scans in behaving mice

A new functional MRI (fMRI) technique cuts down on scanning noise and motion-related artifacts, broadening the range of brain scans possible in awake and behaving animals, according to a new study.

“This is really groundbreaking work,” says Alessandro Gozzi, director of the Functional Neuroimaging Laboratory at the Istituto Italiano di Tecnologia, who was not involved in the study. One of the biggest challenges for scientists doing fMRI in awake animals has been conventional methods’ sensitivity to motion. The degree to which the new method reduces motion artifacts “is really remarkable,” Gozzi says. “The fact that it is silent is just a bonus to me.”

Because animals cannot be instructed to lie still for an fMRI scan, scientists typically anesthetize them first, which limits the types of experiments that can be carried out and raises questions about how sedatives might influence brain activity, says study investigator Yen-Yu Ian Shih, professor of neurology at the University of North Carolina at Chapel Hill.

Similarly, Shih says, stress and other factors may influence scans in awake animals that are acclimated to restraints, due to the noise levels, which can reach more than 100 decibels in conventional fMRI as the scanner sweeps back and forth and rapidly switches magnetic field gradients to acquire blood-oxygen-level dependent (BOLD) contrasts.

The new technique, dubbed “Steady-state On-the-Ramp Detection of INduction-decay with Oversampling” (or SORDINO, the word used to instruct musicians to mute their instruments), captures imaging data by rotating smoothly through space, akin to how the hands of a clock sweep across its face. This method keeps the magnetic field gradient nearly constant while also reducing the time between the radiofrequency pulses generated by the scanner and data acquisition to nearly zero. These findings were published last month in Nature Neuroscience.

This is a “very exciting advance,” particularly for animal studies, says Ravi Menon, professor of medical biophysics, medical imaging, neuroscience and psychiatry at the University of Western Ontario, who was not involved in this work. “They’ve combined a bunch of existing physical principles into a very clever application for functional MRI.” 

And unlike the “direct imaging of neuronal activity” (DIANA) fMRI method that was the subject of intense scrutiny—and, eventually, a retraction—this technique is rooted in well-established physiology that underlies many MRI techniques, including BOLD-based fMRI, Menon adds.

S

SORDINO also reduced electromagnetic interference, making this method more compatible with other techniques, such as electrophysiology. “We also do a lot of electrophysiology, so being able to do that simultaneously with fMRI during social interactions could be extremely exciting,” says Menon, whose team (along with one of the study co-authors, who recently joined Menon’s lab as a postdoctoral researcher) has plans to test this method in marmosets. 

Although SORDINO and BOLD signals both depend on neurovascular coupling, the BOLD signal captures the difference between oxygenated and deoxygenated hemoglobin, whereas the SORDINO signal is linked to cerebral blood volume and blood flow, as well as tissue oxygenation, the study suggests. The ability to detect tissue oxygen is especially exciting, Shih says, because it has the potential to provide a more localized signal than BOLD can. 

But the study didn’t rule out the possibility that the the tissue oxygen the team is detecting is, at least in part, oxygen that is dissolved in blood without being bound to hemoglobin, says Peter Bandettini, chief of the Section on Functional Imaging Methods and director of the Functional Magnetic Resonance Imaging Core Facility at the U.S. National Institute of Mental Health, who was not involved in this work. “It would be great if we were looking at just dissolved oxygen in tissue,” Bandettini adds. “But I think that’s hopeful thinking.” 

Mechanisms aside, Bandettini says that he found the paper convincing, and that this technique has the potential to open new avenues to human fMRI—if it can be successfully adapted. One issue, he says, could be resolution: The study was conducted at high resolution, in which the contribution of noise from physiologic sources such as respiration and heart beats is minimal. Experiments in people are typically carried out at lower levels of spatial resolution, in which these sources of noise can be more troublesome. Whether SORDINO will be able to detect signals in this context remains to be seen. 

Shih says he and his team are working on translating SORDINO for humans, in hopes that it will facilitate fMRI experiments in people who are sensitive to noise or have difficulty lying still in a scanner, such as infants or people with conditions such as autism, PTSD or movement disorders. For now, Shih says that his team, as well as many other researchers at his university, have already switched to using SORDINO almost exclusively for animal fMRI experiments. 

Menon and Gozzi both note that an open question is how well different researchers will be able to implement this method in their labs, and whether this method will be effective in less-powerful MRI scanners. “This is the birth of a new approach, so people will have to play with it and see if it works,” Menon says. “I’m really interested to see how other people use it.”