This week we watched a robot do a backflip, marveled about an implantable device that can help the brain form memories, and learned about an X-ray-reading AI that outperformed Stanford doctors in diagnosing pneumonia. We are starting to feel sorry for sci-fi writers.
What is it? Scientists at the University of Southern California in Los Angeles and Wake Forest School of Medicine in North Carolina are working on a brain implant that could improve memory. They already have tested it in humans suffering from epilepsy, “demonstrating successful implementation of a new neural prosthetic system for the restoration of damaged human memory,” the team reported this week.
Why does it matter? The researchers wrote in a January 2017 paper that such memory “prosthetics” could use implanted circuits to bypass damaged parts of the hippocampus, the area of the brain that encodes memory, and “then stimulate the appropriate effectors, downstream from the site of injury,” i.e. help form memories. The device has “the potential to restore function in diseases such as Alzheimer’s,” the team wrote. The New Scientist wrote that the device could “boost performance on memory tests by up to 30 percent. A similar approach may work for enhancing other brain skills, such as vision or movement,” the magazine reported.
How does it work? The team has been working on the device, called Parylene C, for several years. The device has several probes “anatomically matched to multiple regions of the hippocampus,” a part of the brain that is involved in memory. When the team implanted it into the hippocampus of rats, for example, they were able to “successfully demonstrate recordings from targeted regions” and acquire for the first time “dense neural information from deep brain targets of interest” that play an important role in forming memories.
What is it: Is there anything Boston Dynamics’ Atlas robots cannot do? The latest version of machine has learned to do backflips and other moves previously reserved for gymnasts.
Why does it matter? An older Atlas already went for a walk by itself. The robot, which is 5 feet 9 inches tall and weighs 180 pounds, could also move objects and even stoically handle a few mean pranks by humans (at least for now). The new version could be your gym trainer, an Olympic athlete, you name it. The company, which spun out of MIT, says that it is “combining the principles of dynamic control and balance with sophisticated mechanical designs, cutting-edge electronics, and software for perception, navigation, and intelligence.”
How does it work? You can see them in action yourself, but in a nutshell the electric robots have 28 joints and rely on a battery, lots of hydraulic actuators, LIDAR for navigation and stereo vision, and other technology.
What is it: Computer scientists at Stanford University have written an algorithm that can learn from chest X-ray images and diagnose pneumonia better than radiologists. The code, called CheXNet, reportedly can diagnose up to 14 medical conditions.
Why does it matter? Some 1 million Americans are hospitalized every year with pneumonia, and the disease can be difficult to diagnose. “The motivation behind this work is to have a deep learning model to aid in the interpretation task that could overcome the intrinsic limitations of human perception and bias, and reduce errors,” Matthew Lungren, co-author of the team’s paper, told Stanford News. “More broadly, we believe that a deep learning model for this purpose could improve health care delivery across a wide range of settings.”
How does it work? The team trained the algorithm on more than 100,000 public chest X-ray images released by the National Institutes of Health. Next, they checked the results against four Stanford radiologists’ independent diagnoses. “Within a week the researchers had an algorithm that diagnosed 10 of the pathologies labeled in the X-rays more accurately than previous state-of-the-art results,” Stanford News reported. “In just over a month, their algorithm could beat these standards in all 14 identification tasks. In that short time span, CheXNet also outperformed the four Stanford radiologists in diagnosing pneumonia accurately.”
What is it? Scientists at the Swiss Federal Laboratories for Materials Science and Technology (EMPA) have developed rubber that generates electricity when stretched and compressed.
Why does it matter? The material could lead to miniature rubber power plants that can be worked into buttons and clothing to generate electricity from movement. “This material could probably even be used to obtain energy from the human body,” said EMPA researcher Dorina Opris. “You could implant it near the heart to generate electricity from the heartbeat, for instance.” This could power pacemakers or other implanted devices, eliminating the need for invasive operations to change their batteries.
How does it work? The material takes advantage of the piezoelectric effect allowing certain materials to convert stress and vibrations into electricity. The effect was first discovered in 1880, but application have been mostly limited to stiff materials. “Opris and her colleagues have now succeeded in giving elastomers piezoelectric properties,” EMPA reported.
What is it? Researchers at Sangamo Therapeutics said they edited genes inside a patient’s body for the first time. “For the first time, a patient has received a therapy intended to precisely edit the DNA of cells directly inside the body,” said Dr. Sandy Macrae, CEO of Sangamo Therapeutics. “We are at the start of a new frontier of genomic medicine.”
Why does it matter? The patient in question suffered from Hunter syndrome, a genetic disease caused by a faulty enzyme that leads to “the buildup of massive amounts” of harmful substances that can damage appearance, mental development and other functions, according to Mayo Clinic. There’s no cure for the condition.
How does it work? Sangamo wants to use gene editing “to insert a corrective gene into a precise location in the DNA of liver cells with the goal of enabling a patient’s liver to produce a lifelong and stable supply of an enzyme he or she currently lacks.”