By Kate Becker
Three ENG teams have earned BU Ignition Awards. Conferred annually by BU Technology Development, Ignition Awards are designed to accelerate the advancement of promising new science and technology. Along with financial support, the awards come with coaching and mentoring from industry veterans.
“Ignition Awards are different from other grants,” says Michael Pratt (Questrom’13), Technology Development’s managing director. “They offer a rigorous process through which BU researchers can build new skills, like business proposal development and pitching, meet people who can help them advance their ideas, and ultimately identify a viable pathway to bring a new product to market.”
Processing Encrypted Data in the Cloud
Today, the most private details of our lives—from bank account numbers to medical records—are all in the cloud. Encryption protects that intimate data while it’s stored and while it’s in transit, but generally, the data has to be unencrypted before it can be processed. For example, imagine asking a cloud-based AI like ChatGPT a sensitive question about your health or finances. Your query is at its most vulnerable while it is being processed—in fact, about 64 percent of attacks happen while data is being processed, representing a loss of about $6 billion a year.
Computer scientists have long known that it’s possible to process data while it’s still encrypted, but the technology for doing so is gruelingly slow: it takes 10,000 times longer than standard processing. Fortunately, Rashmi Agrawal (ENG’17,’23), a BU College of Engineering research scientist, discovered a way to process encrypted data some 1,000 times faster. She uses a type of computer chip called a field-programmable gate array, which straddles traditional boundaries between hardware and software by allowing engineers to “program” the circuitry of the chip on the fly.
With Ajay Joshi, an ENG professor of electrical and computer engineering, she formed a company, CipherSonic Labs, to commercialize the technology. So far, they have shown that it works on an academic cloud environment called Mass Open Cloud. Their Ignition Award will allow them to expand the technology to commercial cloud computing platforms like Amazon Web Services.
Monitoring Kidney Dialysis in Real Time
It’s true what you’ve heard: the human body is mostly water. And we have the kidneys to thank for keeping our bodies’ water levels just right. But kidney disease disrupts this natural equilibrium. For roughly half a million Americans with the disease, dialysis can manually restore this fluid balance while removing toxins from the body via the blood.
“Dialysis is a huge lifeline. It’s vital for these patients,” says Darren Roblyer, an ENG associate professor of biomedical engineering and of electrical and computer engineering. “But there’s a problem: it’s not totally clear how much fluid to remove from each patient.” Remove too much water, and the patient can have painful muscle cramps and perilously low blood pressure. Remove too little, and the excess fluid can overload the heart, ultimately leading to heart failure. Today, patients have to rely on trial and error to get the balance right; in roughly two-thirds of cases, dialysis removes too much or not enough fluid.
Now, Roblyer’s lab is developing a device called DialySight that can track a patient’s fluid levels in real time during dialysis. Designed as an anklet, DialySight shines infrared light at a patient’s leg, then measures the reflected light to calculate how much water the tissue contains. In the future, the results could be used to fine-tune the rate and duration of dialysis to make sure that the procedure removes exactly the right amount of fluid.
With support from an Ignition Award, Roblyer has partnered with BU kidney expert Vipul Chitalia, a BU Chobanian & Avedisian School of Medicine professor of medicine, to use the device to take readings from 50 current dialysis patients. The goal: to zero in on key measurements so that he can optimize the DialySight design and make it practical and affordable—and help as many patients as possible.
New Hope for Liver Cancer
Patients diagnosed with the most common type of liver cancer, hepatocellular carcinoma, face a grim prognosis. Most are diagnosed too late or are too sick to qualify for surgery or a transplant. The average patient lives for only 11 months after being diagnosed.
Now, a team of BU biomedical engineers is developing a new way to attack proteins linked with many hepatocellular carcinomas. Their weapon: one of the body’s own proteins, retargeted to break down cancer proteins.
With support from their Ignition Award, Mark Grinstaff, a BU William Fairfield Warren Distinguished Professor, and PhD students Brett Tingley (ENG’25) and Kirk Pierce (ENG’21,’26) aim to show that they can program proteins to home in on cancer-causing proteins so that the body can destroy them. They think this could shrink tumors enough that they can be removed surgically. The approach could also work for other cancer types, including other liver cancers and colorectal cancer.
There are thousands of proteins in the human body that have been linked with diseases. The vast majority of these proteins are undruggable; that is, there are no drugs that target them—yet. Grinstaff and his team hope that their platform could one day be adapted to take on any disease-causing protein.
In total, six BU teams won Ignition Awards.