Welcome to the August 3, 2022, edition of ACM TechNews, providing timely information for IT professionals three times a week.
ACM TechNews mobile apps are available for Android phones and tablets (click here) and for iPhones (click here) and iPads (click here).
To view "Headlines At A Glance," hit the link labeled "Click here to view this online" found at the top of the page in the html version.
The online version now has a button at the top labeled "Show Headlines."
|
|
|
Humanoid Diving Robot Explores Shipwrecks
CNN
Ashley Strickland
July 30, 2022
The OceanOneK robot built by Stanford University researchers is exploring shipwrecks and sunken planes on the ocean floor. The 1.5-meter (five-foot)-long robot resembles a human diver, with arms, hands, and a pair of “eyes” providing three-dimensional vision for full-color representation of the underwater environment for remote operators. Rearward computers and multidirectional thrusters help OceanOneK maneuver around sunken objects. The robot's control system enables operators to sense water resistance and the shapes of artifacts through haptic feedback. The OceanOneK incorporates sensors and algorithms that allow the robot to function autonomously and to avoid collisions.
|
Hardware Offers Faster Computation for AI with Much Less Energy
MIT News
Adam Zewe
July 28, 2022
Researchers at the Massachusetts Institute of Technology (MIT) have developed an analog processor technology that uses inorganic phosphosilicate glass (PSG) to allow for faster computation with greater energy efficiency. Analog machine learning is enabled by increasing and decreasing the electrical conductance of protonic programmable resistors, controlled by the movement of protons into and out of a channel in the resistor. The researchers used PSG to create a programmable protonic resistor that is 1 million times faster than the researchers' previous fastest device; it also can operate at room temperatures using much less energy. Said MIT's Murat Onen, "Once you have an analog processor, you will no longer be training networks everyone else is working on. You will be training networks with unprecedented complexities that no one else can afford to, and therefore vastly outperform them all. In other words, this is not a faster car, this is a spacecraft."
|
Conjoined Twins Separated with the Help of VR
BBC News
Shiona McCallum
August 1, 2022
Virtual reality (VR) technology was instrumental in the surgery that successfully separated Brazilian twins who were joined at the head. The surgery, performed at the Instituto Estadual do Cerebro Paulo Niemeyer in Rio de Janeiro in conjunction with the U.K.'s Great Ormond Street Hospital, involved the use of VR projections of the twins based on computed tomography (CT) and magnetic resonance imaging (MRI) scans. The use of VR technology enabled the surgeons in Brazil and the U.K. to operate on the twins together. Said Dr. Noor ul Owase Jeelani, "In some ways these operations are considered the hardest of our time, and to do it in virtual reality was just really man-on-Mars stuff."
|
3D Printing Process Faster, More Precise Than Conventional Methods
Rutgers Today
Greg Bruno
August 1, 2022
The Multiplexed Fused Filament Fabrication (MF3) method developed by Rutgers University researchers could be used to three-dimensionally (3D)-print large, complex parts at a fraction of the cost of current techniques. MF3 uses a 3D printer's sliding structure or gantry arm to print individual or multiple parts at the same time. The researchers programmed the prototype with slicer software that maps objects into virtual slices and optimizes gantry arm movements for maximum efficiency, while an array of small nozzles, rather than one large nozzle, deposits molten material. The researchers said the new process boosts printing resolution while reducing the time required for printing. Rutgers' Jeremy Cleeman said MF3 "could be a game changer” for the 3D-printing industry.
|
ACM TechBrief Highlights Risks of Quantum Simulation Technology
ACM
August 2, 2022
A technical bulletin from ACM's Technology Policy Council (TPC) weighs the benefits and risks of quantum simulation technologies. The fourth TechBrief to be released by TPC, the bulletin notes media and policymakers' focus on future quantum computers' theoretical encryption-breaking capability has overshadowed quantum simulation's viability and likely impact. In response, the advisory calls for rigorous, immediate planning in order for quantum simulation to fulfill its potential, especially since quantum simulators are likely to emerge before general-purpose quantum computers. ACM TPC chair James Hendler said the TechBrief was intended "to introduce a broad audience to both the promise and peril of quantum simulation so that leaders in science, government, and policy may begin to debate and design parameters within which this technology can develop and be deployed responsibly with maximum benefit and minimum risk."
|
Algorithm Detecting Sepsis Risk Cut Deaths by Nearly 20%
Scientific American
Sophie Bushwick
August 1, 2022
A detection algorithm reduced the hospital mortality rate from sepsis cases by 18% by helping doctors and nurses treat cases nearly two hours earlier on average. Suchi Saria and colleagues at Johns Hopkins University developed the Targeted Real-Time Early Warning System (TREWS), an algorithm that scans patients' electronic health records for factors that compound sepsis risk and combines that information with patients’ vital signs and laboratory tests to produce a score indicating the likelihood they will experience septic shock. Saria's team spent two years incorporating TREWS into the workflow of roughly 2,000 healthcare providers, who used the algorithm in more than 760,000 encounters with patients, including over 17,000 who developed sepsis. TREWS flags a patient's electronic health record to alert providers that the patient is at risk of sepsis, along with reasons why.
|
Technique Protects Data on Solid-State Drives from Radiation
University of Alabama in Huntsville
Jim Steele
August 1, 2022
Researchers at the University of Alabama in Huntsville (UAH) have developed a radiation-resistant computer data storage system. Explained UAH's Biswajit Ray, "We adeptly use the breakdown mechanism of a transistor's oxide layer to imprint information” on commercial off-the-shelf memory cells, a technique that he said was “more resistant to irradiation damage compared to the traditional charge-based technique." The researchers tested chips with as much as 100 times the amount of radiation that would be fatal, “and we find clear benefits using our method compared to the traditional charge based method,” said Ray.
|
Researchers Teaching Computers to Predict Efficient Catalysis
Aarhus University (Denmark)
Peter F. Gammelby
July 28, 2022
A machine learning algorithm developed by researchers at Denmark's Aarhus University and Germany's Fritz Haber Institute can help predict how complex molecules will bind to a catalyst's surface, as well as the strength of the bond. The algorithm also can predict whether a given material could serve as a catalyst in chemical reactions involving numerous complex molecules. Said Aarhus' Mie Andersen said existing algorithms “didn't have the information where the individual atom was located geometrically in the molecule, and how it was bound to the molecule's other atoms and to the atoms on the catalyst's surface.” The new machine learning algorithm, said Andersen, “can extract this information and use it effectively."
|
Post-Quantum Encryption Contender Taken Out by Single-Core PC in One Hour
Ars Technica
Dan Goodin
August 2, 2022
Researchers at Belgium's Katholieke Universiteit Leuven (KU Leuven) ruled out an algorithm selected by the U.S. National Institute of Standards and Technology as a potential post-quantum encryption program. The Supersingular Isogeny Key Encapsulation (SIKE) algorithm was thought to be quantum-decryption-proof by avoiding key encapsulation's vulnerabilities through a supersingular isogeny graph. KU Leuven researchers used a single classical computer to break SIKE, which took it just one hour. The team showed SIKE's linchpin, the Supersingular Isogeny Diffie-Hellman (SIDH) protocol, is vulnerable to a variant of a GPST adaptive attack that "exploits the fact that SIDH has auxiliary points and that the degree of the secret isogeny is known," explained Steven Galbraith at New Zealand's University of Auckland.
|
VegSense Makes Sense for Forest Studies
Rice University News
Mike Williams
August 1, 2022
Rice University bioscientists combined a mixed-reality Microsoft HoloLens headset with the open-source VegSense software to measure vegetation for forest studies. Rice's Daniel Gorczynski said the hardware-software combination is less expensive than light detection and ranging (LiDAR) systems most frequently used in three-dimensional field studies. The headset displays a grid pattern that tracks the surfaces of vegetation as the user scans the environment. Gorczynski said the object "is to get the mesh to cover as much of the vegetation as possible because that's what gets you the best scan." In testing, VegSense detected 48 of 50 relatively mature trees within a circle about 30 feet in diameter in Houston's Memorial Park.
|
Automation Speeds the Search for Stable Proteins
Princeton Engineering News
Stephanie Pappas
August 1, 2022
Researchers from Princeton Engineering and Rutgers University are using a robotic platform to design stable proteins more than 10 times faster than current state-of-the-art methods. Engineers apply a machine learning (ML) model to identify chemical compounds most likely to stabilize desired proteins, winnowing hundreds of thousands of candidates to a few contenders. The robotic assembly platform generates molecular samples, and the platform/ML model combination produces results in as little as a few days. The researchers focused on three proteins with distinctive traits when developing the model; Princeton's Roshan Patel said, "If we could solve the problem for these three, then theoretically we could extend the same procedure to all sorts of enzymes."
|
Ultrasound Stickers Could Continuously Image Internal Organs for Days
New Scientist
Jeremy Hsu
July 28, 2022
Xuanhe Zhao and colleagues at the Massachusetts Institute of Technology have developed a wearable ultrasound sticker that can image the wearer's internal organs continuously for 48 hours. The researchers integrated a transducer that generates and detects ultrasound waves with a patch combining hydrogel to transmit the waves, and flexible elastomer material. The researchers stuck the stickers to volunteers who performed various actions, and were able to image changes in the size and shape of their organs during those activities. The stickers currently have to be connected by wires to a computer, which translates the ultrasound waves into images and collects the data.
|
Computer Modeling Aims to Inform Restoration, Conservation of Coral Reefs
UBC Okanagan News (Canada)
August 2, 2022
Researchers at Canada's University of British Columbia, Okanagan (UBCO) used computer modeling to help scientists forecast how climate damage and restoration approaches could impact coral reefs. The researchers modeled 245 coral communities, each possessing nine unique species and covering a surface of 25 square meters (269 square feet). The team modeled situations like strong waves, cyclones, or intense heat, then quantified each reef's resilience by noting damage, recovery time, and habitat quality 10 years later. More-diverse communities were found to have maximum resilience, while less-diverse reefs were more vulnerable to predatory algae. UBCO's Bruno Carturan said the model "could be used to experiment and perhaps determine if losing these species can be compensated by some other, more resistant ones, that would prevent the eventual collapse of the reefs."
|
|
