A deep learning-based model estimates attenuation maps directly from emission data, enhancing myocardial perfusion imaging on SPECT-only scanners.
An AI trained to spot objects hidden against a background is able to see through camouflage and outperforms existing algorithms at the task.
Almost three months after arriving in Guam, a pair of MQ-4C Triton autonomous, unmanned aircraft have integrated into fleet operations and training flights and stretched the Navy’s maritime domain awareness across the Indo-Pacific, according to the Navy.
The Navy is counting on the Triton, which can operate at greater than 50,000-foot altitudes and at the 2,000-mile-plus range, to provide an unmanned platform for persistent, maritime intelligence, surveillance, and reconnaissance capabilities and work alongside its manned fleet of reconnaissance and surveillance patrol aircraft. The Tritons with Unmanned Patrol Squadron 19 – the Navy’s first unmanned aircraft squadron – arrived in Guam in late January to support CTF-72, which oversees the patrol, reconnaissance and surveillance force in the U.S. 7th Fleet region.
“Bringing Triton forward creates a complex problem set for our adversaries,” Cmdr. Michael Minervini, VUP-19’s commanding officer, said in a statement. “Our ability to provide persistent ISR to fleet and combatant commanders is unmatched in naval aviation.”
https://www.youtube.com/watch?v=eNyRM0c-btw&feature=youtu.be
face_with_colon_three circa 2019.
Robot gives this man a haircut.👦
Continue reading “Robot programmed to give people haircuts” »
Washington (AFP) — Facebook unveiled an initiative Tuesday to take on “hateful memes” by using artificial intelligence, backed by crowd sourcing, to identify maliciously motivated posts.
The leading social network said it had already created a database of 10,000 memes — images often blended with text to deliver a specific message — as part of a ramped-up effort against hate speech.
“These efforts will spur the broader AI research community to test new methods, compare their work, and benchmark their results in order to accelerate work on detecting multimodal hate speech,” Facebook said in a blog post.
Stroke is the leading cause of serious long-term disability in the US with approximately 17 million individuals experiencing it each year. About 8 out of 10 stroke survivors suffer from “hemiparesis”, a paralysis that typically impacts the limbs and facial muscles on one side of their bodies, and often causes severe difficulties walking, a loss of balance with an increased risk of falling, as well as muscle fatigue that quickly sets in during exertions. Oftentimes, these impairments also make it impossible for them to perform basic everyday activities.
To allow stroke patients to recover, many rehabilitation centers have looked to robotic exoskeletons. But although there are now a range of exciting devices that are enabling people to walk again who initially were utterly unable to do so, there remains significant active research trying to understand how to best apply wearable robotics for rehabilitation after stroke. Despite the promise, recent clinical practice guidelines now even recommend against the use of robotic therapies when the goal is to improve walking speed or distance.
In 2017, a multidisciplinary team of mechanical and electrical engineers, apparel designers, and neurorehabilitation experts at Harvard’s Wyss Institute for Biologically Inspired Engineering and John A. Paulson School of Engineering and Applied Sciences (SEAS), and Boston University’s (BU) College of Health & Rehabilitation Sciences: Sargent College showed that an ankle-assisting soft robotic exosuit, tethered to an external battery and motor, was able to significantly improve biomechanical gait functions in stroke patients when worn while walking on a treadmill. The cross-institutional and cross-disciplinary team effort was led by Wyss faculty members Conor Walsh, Ph.D. and Lou Awad, P.T., D.P.T., Ph.D, together with Terry Ellis, Ph.D., P.T., N.C.S. from BU.
In the movie “Transformers,” cars morph into robots, jets or a variety of machinery. A similar concept inspired a group of researchers to combine gas foaming, which is a blend of chemicals that induces gas bubbling, and 3D molding technologies to quickly transform electrospun membranes into complex 3D shapes for biomedical applications.
In Applied Physics Reviews, the group reports on its new approach that demonstrates significant improvements in speed and quality compared with other methods. The work is also the first successful demonstration of formation of 3D neural tissue constructs with an ordered structure through differentiation of human neural progenitor/stem cells on these transformed 3D nanofiber scaffolds.
“Electrospinning is a technology to produce nanofiber membranes,” said co-author Jingwei Xie, at the University of Nebraska Medical Center. “The physics principle behind it involves applying an electrical force to overcome the surface tension of a solution to elongate a solution jet into continuous and ultrafine fibers after solvent evaporation.”
No one has ever seen any airplane like this, except on computer animation. Now, some of the world’s top aeronautical engineers are going to build it for real.
The plan calls for constructing a six-ton unmanned, remote controlled plane the size of a business jet with 24 spinning propellers embedded in its huge moveable wings that allow it to magically hover in midair.
It’s an experimental airplane they call LightningStrike.
A growing cadre of startups is pursuing iterative cycles of machine learning, wet-lab experimentation and human feedback to accelerate target drug discovery.
