BERKELEY, Calif. - She's a physicist who trained in the storied lab where Watson and Crick worked out the structure of DNA. In her years in industry, she made sharper displays for e-readers, more efficient solar panels, and sensor tape that soldiers could wear on the battlefield to measure the strength of explosions.
Her manufacturing tool of choice: a simple printer.
Ana Claudia Arias is an expert in the field of low-cost printable electronics. Now at the University of California, Berkeley, she's focused on using printers loaded with a variety of high-tech inks to make a new generation of medical devices, from wearables to barely noticeable MRI hardware for kids.
"Our dream is to have electronics in things like this," she said, holding out a piece of plastic mesh so soft it felt like cloth.
Arias doesn't wear a Fitbit. Why would she? To her, they're bulky, unattractive and, most annoyingly, have to be recharged all the time.
Instead, in a lab filled with printers that can extrude liquid silver nanoparticles, carbon nanotubes and semiconducting plastics, Arias and her students have created whisper-thin, flexible devices that monitor a variety of body functions.
There's a pulse oximeter the size of a Band-Aid, which reads blood oxygen levels from anywhere on the body and could easily be added to a wrist-worn fitness tracker. Tests show it functions as well as the more expensive rigid ones used in hospitals, which work only on translucent parts of the body such as fingers or earlobes.
To solve the recharging problem, she's working on flexible energy storage devices; in one bracelet that can be charged by the sun or indoor light, the solar cell is crafted to look like a jewel. Running out of power is merely an inconvenience for a fitness tracker, but it's a major issue for monitoring devices in medical settings, said Arias, 44, a native Brazilian.
While the world of medical wearables may be eager for much of the technology Arias is cooking up, she's somewhat exasperated with products for the consumer market. In addition to being too large and uncomfortable for her taste, they're often unreliable. "They don't have to go through FDA approval," she said. "With the Fitbit, sometimes you don't even have a pulse."
Instead, Arias is building devices that, she said, could provide "real, valid medical information caregivers can use." She's working on adapting her pulse oximeter into a bandage-like sensor that could monitor how a wound is healing. She'd like to develop sensors that could slip onto the insoles of shoes to warn diabetics of foot ulcers they can't feel.
"What would be best would be electronics that were almost like a second skin," she said. "No adhesive. No straps. Almost like underwear - you forget that you're wearing it."
A sensor you can barely feel is exactly what Miki Lustig was looking for. Lustig, who works two doors away from Arias, is an electrical engineer and expert in MRI signal processing. He's been working for 15 years on ways to make MRI images crisper and exams shorter.
One of the main problems for all patients, but especially for children, Lustig said, is that traditional MRI coils, which serve as antennae to receive signals from a patient's body, are rigid, big and bulky. They're uncomfortable and often lie too far away from the anatomy that requires imaging, resulting in poor image quality. Some coils are so heavy they can crush a small child's chest and need to be propped up with blankets, pushing them even farther away.
Lustig had been pondering the coil problem when he attended a research talk by Arias, who had joined the Berkeley faculty just weeks before. As she described her flexible electronics, he immediately saw the potential and rushed over as soon as her talk ended.
"He said, 'Ana! Do you think you can print coils?'" Arias recalled. "I said, 'Miki! What are coils?'"
But once she learned about MRI coils - and saw the huge need for better ones in pediatrics - she was hooked. "Immediately we started getting materials and trying to test them," Lustig said.
It was a funny collaboration: Lustig, a software whiz, had never built hardware. Arias could build most anything but didn't know the first thing about MRI. Neither knew anything about radio frequency, a critical component of MRI coils. They forged ahead anyway. "It was a learning curve," Lustig said. Getting to where they are today - testing lightweight, plastic coils on patients at Stanford's Lucile Packard Children's Hospital - took six years.
At the hospital, they worked with Dr. Shreyas Vasanawala, a Stanford pediatric radiologist and engineer. He had grown increasingly frustrated with the barriers that prevented many of his young patients from receiving MRIs.
When children did get these scans, they often had to be anesthetized to ensure they stayed still during the long exams. And sometimes a tube had to be inserted down their windpipe to ensure they could keep breathing under the heavy weight of the coils. "Something that's otherwise a completely safe, noninvasive imaging test becomes a much bigger deal," Vasanawala said.
Physicians and parents often opt for simpler CT scans instead. But those scans involve radiation and also don't image soft tissue nearly as well as MRIs. "Pediatricians have to accept a sub-optimal test," he said.
MRI imaging for children is such a problem, the National Institutes of Health has made funding research in the area - including Vasanawala's and Arias's work - a priority.
"We know MRI can do many wonderful things for adults, but children have been left behind," said Guoying Liu, who directs the MRI program at the National Institute of Biomedical Imaging and Bioengineering and has funded more than a dozen projects that focus on younger patients. "When I first arrived here eight years ago, there was zero technology development for children. There was nothing. I was shocked."
Many credit Vasanawala for being the driving force behind improvements to pediatric MRI. He's spent the last decade working with Lustig and others to find ways to shorten exam times and reduce the need for children to be anesthetized. His team has created software that speeds image gathering and corrects for motion so kids don't have to keep perfectly still. An MRI can now be done in as little as five or 10 minutes - down from more than an hour - Vasanawala said.
His team also created kid-sized flexible coils that are light but denser so more information can be collected for images in a shorter time. These coils are now being developed commercially by GE Healthcare, but are still a bit bulky and definitely noticeable to a child.
Arias's coils, in contrast, are thin, flat plastic pieces with the receiver material - silver nanoparticles - screen-printed on them. "Just like T-shirts," Arias said. The coils are light, flexible, and see-through, like a thick report cover.
When Joseph Corea, the graduate student who constructed the first coils, showed them to Vasanawala, he was worried they might send out a corrupted signal when they were flexed around a body part. But there were no problems of that sort, Vasanawala said, and a host of benefits. They don't press down on children's lungs, for one thing.
"They're so flexible and so low-profile, people don't notice they're there," he said. "MRI techs and nurses tend to focus on the patient instead of the technology."
The crisper medical imaging these coils enable can have a huge clinical impact. Vasanawala described one dramatic case - involving a 5-year-old named Finn Green who was set to have a liver transplant due to cancer. But an MRI scan - with no anesthesia - showed only part of Green's liver was affected, and allowed a surgeon to take out only the cancerous portion of the liver. The child's liver grew back to normal size in six weeks. He remains healthy and cancer-free.
Lighter and flexible coils are the future, said Jason Polzin, who leads MRI technology development for GE Healthcare and is working with Arias and Vasanawala to incorporate their technology into next-generation coils. "We ask ourselves, how did it go this long with the ways coils are now?," he said. "It seems so obvious in hindsight."
To speed development of their flexible MRI coils, Arias, with Lustig, Corea, and Balthazar Lechene, a French postdoctoral researcher and materials scientist, have spun off a company called InkSpace Imaging. The name refers to ink for printing, of course, but also to k-space - a technical term for an array of numbers that represents spatial frequency in an MRI image.
"It's very geeky," Arias said. "We get a laugh every time we think of it."
To get the coils to work in a pediatric hospital, however, also required skills Arias's team didn't learn in engineering school.
They had to find soft, fuzzy fabric for a "blankie" in which to place the coils. The team went to a Joann store and promptly started arguing over which patterns and colors to use and whether they were gender-neutral or comforting enough. They settled on a bright green, fire-resistant fabric with extremely cute dinosaurs saying 'RAWR.'
"We were asking things like, 'Would 12-year-olds like dinosaurs?'," recalled Lechene, who helped develop the coils and now keeps bolts of dinosaur cloth in his office. "These are not questions we normally think about."
There was just one last hurdle: Arias had to teach her engineering students how to sew.