BUILDING IMMORTAL MACHINES
Many electricity-generating turbines have an official life of 30 years, yet some of the machines that GE maintains have been operating for twice as long — one is still running after 82 years. Colin Parris, vice president of software research at GE’s Global Research Center, is taking the kind of life extension that occurs by happenstance and turning it into a formal engineering practice. The idea is to make more efficient machines that last forever.
The forever turbine: The project involves combining three important technologies: computer simulation, artificial intelligence and 3D printing. For instance, if a part fails and needs to be replaced, GE engineers could use the opportunity to design and 3D-print a better, more highly customized component that enables the machine to perform better. The engineers could also run simulations on virtual versions of the turbines called digital twins, which use machine learning to predict how the turbine will function in the years to come. “Normally a model will describe how a turbine works, but every machine degrades in different ways depending on a variety [of] factors,” Parris said.
Read more about the GRC project here.
A PICTURE-PERFECT PARTNERSHIP
GE and the University of Wisconsin have found many a success in improving imaging technology throughout their three decades long partnership. For instance, GE placed a “pre-product” version of their latest MR scanner, the SIGNA Premier, at UW. That let university researchers conduct neurological, liver, body, cardiac, prostate and breast MR studies on real patients, feats that are challenging for GE in a factory setting.
Beyond pretty pictures: In June 2018, GE installed a second SIGNA Premier system at the University of Wisconsin to be used solely for research. The scanner is wider, so researchers can, for example, measure liver fat in obese patients and study blood flow to better understand how the placenta develops during pregnancy. Another group of researchers will use the machine to study the causes of Alzheimer’s disease and potential treatments by repeatedly scanning patients and monitoring changes. GE and UW are also discussing how to best use PET/MR technology in clinical applications and how artificial intelligence can further improve imaging.
Read more about the imaging advances from GE and UW here.
GROUNDBREAKING HUMAN HEALTH DATA PROJECT
Earlier this year, Google’s health research company Verily announced a collaboration with Duke University and Stanford University called Project Baseline. The researchers plan to map human health data, collected from 10,000 people over four years, and present it through an intuitive platform much like Google Earth.
A body of knowledge: Project Baseline’s goal is to learn how specific diseases, such as certain kinds of cancer, relate to particular populations, and then try to prevent them. The difference between this epidemiological study and previous ones is the immense computing power and machine learning capabilities that Google can provide. Since its launch, roughly 2,000 people have signed up and the team has collected roughly 6,700 terabytes of data. The diversity of the participants could also help the researchers to find different risk factors among different ethnicities.
Read more about Project Baseline here.
— VIDEO OF THE WEEK —
— QUOTE OF THE DAY —
“By working together, you’re getting the best of both sides. And that’s what advances medicine and provides the best possible technology for patients.”
— Frank Korosec, professor of radiology and section chief of Imaging Sciences in the Department of Radiology at UW
Quote: GE Reports. Image: Getty Images.
ENJOY THIS NEWSLETTER?
Please send it to your friends and let them know they can subscribe here."