Latest Guides

Science and Technology

Caltech: Good Vibrations

Published on Tuesday, July 21, 2015 | 11:10 am
Michael Shapiro (left) and Doris Tsao. Photo by HC Van Urfalian

For their 1966 song, “Good Vibrations,” the Beach Boys assembled an unusual mix of instruments—including a jaw harp, a cello, and an Electro-Theremin—to produce one of their biggest hits. By arranging sound waves in a unique and particular way, they were able to elicit a positive response.

Many doctors and researchers have the same goal. After all, the same “excitations” that helped the Beach Boys usher in an era of feel-good pop—the sound waves that propagate through air and water, bringing notes of music to our ears—are also noninvasively able to explore body tissues, helping to visualize babies in the womb, heal back pain, or even deliver chemotherapeutics to targeted tumors.

One of the most-used “good vibrations” in medicine is ultrasound—sound waves delivered at a frequency inaudible to the human ear. Ultrasound has been used in medical settings since the 1940s for diagnostics and in recent years has gained popularity for use in physical therapy and to speed up drug delivery.

But that, say chemical engineer Mikhail Shapiro and biologist Doris Tsao, isn’t all that ultrasound can do. The two, who met shortly after Shapiro was hired to the Caltech faculty in late 2013, have joined forces to develop a new technology that uses ultrasound to both map and determine the function of interconnected brain networks. Their goal: to one day be able to change abnormal neural activity deep within the brain using pulses of sound. The idea is so intriguing that, in September 2014, theirs became one of 58 projects nationwide to be awarded funding by the National Institutes of Health (NIH) as part of President Obama’s Brain Research through Advancing Innovative Neurotechnology—or BRAIN—Initiative.

They are an ideal pair for this project: Shapiro’s lab focuses on ways to use different forms of energy—like magnetic fields or sound waves—to penetrate deep into the brain in order to image or control specific processes, like neural function. Tsao, for her part, works to pinpoint specific areas of the brain where functions such as object perception occur. Together—and with the help of postdoctoral fellows Jerome Lacroix in Shapiro’s lab and Tomo Sato in Tsao’s lab—they hope to combine their specialties to use sound waves to inhibit or excite different areas of the brain in order to obtain a specific response.

Their idea plays off of a technique called deep brain stimulation (DBS), which uses implanted electrodes to send electrical impulses to tightly targeted regions of the brain; those impulses block abnormal nerve signals to address severe, treatment-resistant depression and epilepsy, among other movement and affective disorders. The problem is that, as you might imagine, the technique requires highly invasive surgery, during portions of which the device’s recipient needs to be awake.

“If you could stimulate the regions involved in such conditions in a noninvasive way—with ultrasound waves, for instance—it would be a huge advantage,” Tsao says.

They have reason to believe that’s possible. For one thing, scientists at other institutions—including neurobiologist Jamie Tyler at Arizona State University—have shown that you can use ultrasound to stimulate brain cells in rodent models.

“For example, Tyler showed you could make a mouse flick its tail when certain parts of its brain were stimulated in the motor cortex,” says Tsao, whose introduction to Tyler at a meeting a few years ago inspired her to start her own investigations into controlling brain cells with ultrasound waves. “It became obvious to me that if this could work in humans, it would have tremendous impact. With ultrasonic neuromodulation, not only could we stimulate any part of the human brain noninvasively, but we could ask subjects about their experience and do it all inside an MRI scanner, the data from which we could use to map the connectivity and gain a greater understanding of how the brain functions.”

The problem is that the mechanism for how the neurons become excited or inhibited by ultrasound is largely a question mark. In fact, at the most basic level, no one quite knows how DBS works, either. So Shapiro and Tsao are, to some degree, starting from scratch.

To begin, the team wants to study precisely what happens at the cellular and molecular level when ultrasound waves come into contact with certain neurons. So they have built an experimental setup with which they can use microscopy and electrophysiology techniques to look at what’s happening to cells and molecules while they are being bombarded by ultrasound waves. “You need a way for the sound waves to have more or less unfettered access to your cells,” explains Shapiro.

Their solution was a big water tank into which an ultrasound transducer is submerged; brain cells, grown on a nutrient gel, are then placed on the surface of the water. Shapiro and Tsao can look at the cells from above with a normal microscope; they also have an electrophysiology device in contact with the cells to measure electrical activity in the neurons. “Constructing this exotic setup was the first step, and we’re there,” Shapiro says.

Next, they want to use this setup to figure out what mechanism excites brain cells when they’re hit with ultrasound. From a technical point of view, they also want to figure out the limits of the technology and how to optimize it for use in different types of animals, with the goal of eventually testing it in humans.

“This technology has a long way to go,” Shapiro notes. “But if at the conclusion of our three-year grant we’ve achieved all of our goals, we’ll be in a really great position to expand our research, maybe even into human trials.”

Tsao, for her part, is already looking even further down the road. “I’d like to be able to pass this technique on to people at Caltech like John O’Doherty and Antonio Rangel who could potentially use it in their work on behaviors like addiction that are regulated by the brain” she says. “So if it does work, there are a large number of people who are at the ready to translate this. And that’s really exciting.”


Finding a natural way to power those deep-brain stimulators once they are buried in the brains of people—as well as other devices implanted in or near the head—is the task electrical engineer Hyuck Choo has set himself.

“If you use a battery, you have to replace it at some point,” he explains. “And if the device is already in the body, that means you have to have another surgical procedure. This makes people hesitant to use implants. If you can have a more permanent power source that continuously powers the devices, it would be a big advantage. So I’ve been looking for an energy source that we could reliably and easily harvest, or capture and store.”

Choo thinks he’s found a solution, one we use all the time: our voice. He wants to harness the vibrations that vocal cords make when we talk, and use them to power implantable devices. So, for example, a deaf person could sing a song to charge up their cochlear implant.

Last summer, Choo tasked undergraduate Sophia Chen with testing this idea—that our voices could be used to power devices—as part of her Summer Undergraduate Research Fellowship (SURF). “First, I analyzed the vocal-cord vibrations throughout the skull, which basically showed that we could harness those vibrations and turn them into storable energy,” explains Chen.

Sophia Chen attaching sensors to the face of a volunteer. Photo by Lance Hayashida

To then test the strength of the vibrations, she simply attached tiny accelerometers to different areas of the head in 8 volunteer participants, one of whom was Choo. She asked the volunteers to hum at a constant frequency; then hum on a scale, from lowest to highest frequency; and then read aloud.

“Sophia found that, no matter the vocal activity being performed, the acousto-mechanical vibrations were concentrated at a single frequency for 80 to 90 percent of the time,” says Choo. “By focusing on one frequency, we can really optimize the harvesting process.”

This suggests that—instead of having wires running from the brain to a power source typically placed under the skin of your chest for deep-brain stimulation, or batteries mounted behind the ear for cochlear implants—a device harvesting energy from vocal cords could also be implantable. “There would be no wires sticking out or anything; everything would be inside the head,” says Choo. “That would be an advantage of this approach.”

He is now working on building such a vibration-harnessing device, inspired by an off-the-shelf piezo-electric setup—a device that harvests energy from pressure, including that derived from sound vibrations. The team’s challenge is to scale down that technology so it can be implanted in the body while leaving it sizeable enough to generate the power needed, utilizing the energy provided. “Best we can tell right now from the data we have, a person would have to sing their favorite songs for about 10 to 20 minutes twice a day to keep their device powered,” says Choo.

He notes that there is no worry of overdoing it, should one want to sing an entire opera or gab with friends for hours. A safety feature can be built into the device to cut off the charging process once the implant has enough juice.

Because Chen—now finishing up her sophomore year at Caltech— has coursework to deal with, Choo and his lab are taking the SURF work she did and running with it. But Choo gives Chen all the credit for what he calls “a very viable option” for solving this medical challenge.

“The right project for the right student makes a big difference,” says Choo. “We are continuously working on this project, and when the time comes to test the energy-harvesting device that we fabricate, I hope Sophia will come back and help us again.”


Ultrasound machines and energy harvesters use regular old sound waves in unique and novel ways. But Caltech senior postdoctoral scholar Carly Donahue has taken a different tack; she helped devise techniques to try to change the way those sound waves behave, with the goal of giving sound more power in medical applications.

In a research group lead by Chiara Daraio, Donahue and graduate student Paul Anzel worked to produce highly focused, high-amplitude acoustic signals called sound bullets because of their destructive power. The manipulated sound waves act much like a tidal wave, appearing to move by pushing all their energy forward in a single crest instead of in the classic squiggly (and weaker) waveform. The hope is that these focused packets of energy could one day be used to destroy unwanted tissue or trim away tumors, all without doing the kind of damage to the body that a real bullet would do.

This work really began in 2010, when Daraio and her lab reported that they had learned how to control the way in which sound travels, using granular materials—in their case, macroscopic stainless-steel ball bearings, or spheres, assembled into a chain. An array of 21 such parallel chains created what the researchers call an acoustic lens—a pulse of sound pressure initiated at one end travels down each chain in much the same way motion in the Newton’s cradle children’s toy travels from the first ball to the last in a chain without moving the ones in the middle. The point of it all? To use the lens to focus all that pressure, all that sound, at one spot, creating a sort of bullet of sound.

“It’s a simple concept, but it has such incredibly interesting physics,” Donahue says. “The whole idea of the lens is being able to control acoustic signals in a completely different, non-linear way.” According to Anzel, who is studying applied physics, the best way to focus sound at a specific point is to shape the way the signal moves through space.

“What we took advantage of is that, just by squeezing a row of bearings, you can cause a signal to travel through it faster,” he says. “So, if you create rows of ball bearings and squeeze the outer rows more than the inner ones, you can control the speed of the sound pulses so that they arrive at the same time from a bunch of different directions to target an area.”

Building on the 2010 results, which focused the sound waves to penetrate a solid material, Donahue and a team of colleagues began testing the concept in liquid. They also began thinking about applications: sound bullets traveling through liquids could be used to image and evaluate the structures of bridges, says Anzel, or solid objects on the floor of the ocean, much like weaker ultrasound waves are used to image the body.

“We wanted to try this in liquid because, if we want to do medical applications, we have to deal with the fact that obviously the body is not solid,” Donahue explains. “For liquid, we had to think about how to actually transmit a wave from a solid material into a fluid, and the extra complications involved.”

To create the lens in water, Donahue, Daraio, Anzel, and others first aligned the same spheres they’d used in their solid lens, but put each chain into individual tubes. Then, they constructed a waterproof interface—made of a glass disk and polymer encasing—so that they could submerge part of the tubes in water without the spheres falling out. After arranging the tubes next to each other to form the same kind of array they’re created in the solid version of the lens, they then generated a sound pulse, controlling each wave’s timing and amplitude.

“The most surprising part of the study was how important the materials used in the lens-water interface were for controlling how the sound travels,” Daraio says. “We knew it would play a substantial role in the formation of the sound bullet and the energy trans- fer to the water, but we had not fully realized the complexity required in its design. This is something Carly really made important progress on.”

A much smaller setup would be needed for use in the human body, so Daraio’s research group is investigating solutions for the miniaturization of such a device. And Donahue is now exploring how contacts between different materials work on smaller scales—at the microscale or even the nanoscale—where different forces may come into play. “We want to know the basic forces and how things behave to see if it will work,” she says.

If successful, these tiniest of sound bullets might one day be used to noninvasively blast kidney stones and gallstones, remove blood clots, and potentially provide a more accurate imaging alternative, one that could produce even clearer views of the body than do current ultrasound technologies.

“Reducing the device’s size would enable us to reach wavelengths of interest for medical applications, and I hope to see this realized in the next few years,” says Daraio. “It is exciting to see many years of hard work and passion for fundamental research leading to the creation of an instrument that may improve everyday life for many people.”


Get our daily Pasadena newspaper in your email box. Free.

Get all the latest Pasadena news, more than 10 fresh stories daily, 7 days a week at 7 a.m.

Make a comment

Your email address will not be published. Required fields are marked *