Now in 3D: Deep learning techniques help visualize X-ray data in three dimensions

A team of Argonne scientists has leveraged artificial intelligence to train computers to keep up with the massive amounts of X-ray data taken at the Advanced Photon Source.

Department of Energy awards $4.15 million to Argonne to support collaborations with industry

The U.S. Department of Energy has awarded $4.15 million to Argonne National Laboratory to support collaborations with industry aimed at commercializing promising energy technologies.

Novel Method Predicts if COVID-19 Clinical Trials Will Fail or Succeed

Researchers are the first to model COVID-19 completion versus cessation in clinical trials using machine learning algorithms and ensemble learning. They collected 4,441 COVID-19 trials from to build a testbed with 693 dimensional features created to represent each clinical trial. These computational methods can predict whether a COVID-19 clinical trial will be completed or terminated, withdrawn or suspended. Stakeholders can leverage the predictions to plan resources, reduce costs, and minimize the time of the clinical study.

Liquid Metal Sensors and AI Could Help Prosthetic Hands to ‘Feel’

Prosthetics currently lack the sensation of “touch.” To enable a more natural feeling prosthetic hand interface, researchers are the first to incorporate stretchable tactile sensors using liquid metal and machine learning. This hierarchical multi-finger tactile sensation integration could provide a higher level of intelligence for artificial hands by improving control, providing haptic feedback and reconnecting amputees to a previously severed sense of touch.

SLAC hosts Secretary of Energy Jennifer Granholm for a virtual visit

Highlights of the two-hour visit included behind-the-scenes looks at one of the most powerful X-ray sources on the planet and at the construction of the world’s largest digital camera for astronomy. She also joined presentations of the lab’s research in machine learning, quantum technology and climate science and engaged in discussions about diversity, equity and inclusion at SLAC.

Using Computation to Improve Words: Model Offers Novel Tool for Improving Serious Illness Conversations

Conversations between seriously ill people, their families and palliative care specialists lead to better quality-of-life. Understanding what happens during these conversations – and how they vary by cultural, clinical, and situational contexts – is essential to guide healthcare communication improvement efforts. To gain true understanding, new methods to study conversations in large, inclusive, and multi-site epidemiological studies are required. A new computer model offers an automated and valid tool for such large-scale scientific analyses.

AI Learns to Predict Human Behavior from Videos

New Columbia Engineering study unveils a computer vision technique for giving machines a more intuitive sense for what will happen next by leveraging higher-level associations between people, animals, and objects.“Our algorithm is a step toward machines being able to make better predictions about human behavior, and thus better coordinate their actions with ours,” said Computer Science Professor Carl Vondrick. “Our results open a number of possibilities for human-robot collaboration, autonomous vehicles, and assistive technology.”

AI Predicts How Patients with Viral Infections, Including COVID-19, Will Fare

UC San Diego School of Medicine researchers discovered gene expression patterns associated with pandemic viral infections, providing a map to help define patients’ immune responses, measure disease severity, predict outcomes and test therapies — for current and future pandemics.

Machine Learning System Improves Accelerator Diagnostics

A machine learning system is helping operators resolve routine faults at the Continuous Electron Beam Accelerator Facility (CEBAF). The system monitors the accelerator cavities, where faults can trip off the CEBAF. The system identified which cavities were tripping off about 85% of the time and identified the type of fault about 78% of the time.

Medical AI models rely on ‘shortcuts’ that could lead to misdiagnosis of COVID-19 and other diseases, UW researchers find

University of Washington researchers discovered that AI models ignored clinically significant indicators on X-rays and relied instead on characteristics such as text markers or patient positioning that were specific to each dataset to predict whether someone had COVID-19.

DOE names six Argonne scientists to receive Early Career Research Program awards

Six Argonne scientists receive Department of Energy’s Early Career Research Program Awards.

Argonne researchers using artificial intelligence to shape the future of science

Artificial intelligence is being called “the next generation of the way we do science.” At Argonne, researchers are leveraging the lab’s state-of-the-art-facilities and unparalleled expertise to shape the very future of science.

ORNL’s Sergei Kalinin elected Fellow of the Microscopy Society of America

Sergei Kalinin, a scientist and inventor at the Department of Energy’s Oak Ridge National Laboratory, has been elected a Fellow of the Microscopy Society of America professional society.

Catastrophic Sea-Level Rise from Antarctic Melting is Possible with Severe Global Warming

The Antarctic ice sheet is much less likely to become unstable and cause dramatic sea-level rise in upcoming centuries if the world follows policies that keep global warming below a key 2015 Paris climate agreement target, according to a Rutgers coauthored study. But if global warming exceeds the target – 2 degrees Celsius (3.6 degrees Fahrenheit) – the risk of ice shelves around the ice sheet’s perimeter melting would increase significantly, and their collapse would trigger rapid Antarctic melting. That would result in at least 0.07 inches of global average sea-level rise a year in 2060 and beyond, according to the study in the journal Nature.

Helping companies use high-performance computing to improve U.S. manufacturing

Argonne is helping U.S. companies solve pressing manufacturing challenges through an innovative program that provides access to Argonne’s world-class computing resources and technical expertise.

ORNL’s superb materials expertise, data and AI tools propel progress

At the Department of Energy’s Oak Ridge National Laboratory, scientists use artificial intelligence, or AI, to accelerate the discovery and development of materials for energy and information technologies.

CHOP Researchers Demonstrate How Dynamic Changes in Early Childhood Development May Lead to Changes in Autism Diagnosis

Researchers found that difficulties in diagnosing toddlers with autism spectrum disorder (ASD) might be due to the dynamic nature of the disorder during child development. Children with clinical characteristics that put them on the diagnostic border of autism have an increased susceptibility to gaining or losing that diagnosis at later ages.

New rapid COVID-19 test the result of university-industry partnership

A partnership between UC Davis and Maurice J. Gallagher, Jr., chairman and CEO of Allegiant Travel Company, has led to a 20-minute COVID-19 test. The method pairs a mass spectrometer with a powerful machine-learning platform to detect SARS-CoV-2 in nasal swabs. A recent study published in Nature Scientific Reports shows the test to be 98.3% accurate for positive COVID-19 tests and 96% for negative tests.

New Argonne partnership to predict fuel injector dynamics

Collaborators use experiments, high-fidelity simulations and machine learning to deliver predictive tools to engine manufacturers.

8 Things Argonne is Doing to Save the Earth

Stepping into their superhero gear, Argonne scientists are using science and the world’s best technology to combat some of Earth’s toughest foes, from pollution to climate change.

Software Package Enables Deeper Understanding of Cancer Immune Responses

Researchers at the Bloomberg Kimmel Institute for Caner Immunotherapy at the Johns Hopkins Kimmel Cancer Center have developed DeepTCR, a software package that employs deep-learning algorithms to analyze T-cell receptor (TCR) sequencing data. T-cell receptors are found on the surface of immune T cells. These receptors bind to certain antigens, or proteins, found on abnormal cells, such as cancer cells and cells infected with a virus or bacteria, to guide the T cells to attack and destroy the affected cells.

April Snapshots

Science Snapshots from Berkeley Lab: X-rays accelerate battery R&D; infrared microscopy goes off grid; substrates support 2D tech

Game on: Science Edition

UPTON, NY — Inspired by the mastery of artificial intelligence (AI) over games like Go and Super Mario, scientists at the National Synchrotron Light Source II (NSLS-II) trained an AI agent — an autonomous computational program that observes and acts — how to conduct research experiments at superhuman levels by using the same approach. The Brookhaven team published their findings in the journal Machine Learning: Science and Technology and implemented the AI agent as part of the research capabilities at NSLS-II.

Argonne’s 2021 Maria Goeppert Mayer Fellows bring new energy, promise to their fields

The Department of Energy’s Argonne National Laboratory is proud to welcome five new FY21 Maria Goeppert Mayer Fellows to campus, each chosen for their incredible promise in their respective fields.

New machine learning tool diagnoses electron beams in an efficient, non-invasive way

For the past few years, researchers at the Department of Energy’s SLAC National Accelerator Laboratory have been developing “virtual diagnostics” that use machine learning to obtain crucial information about electron beam quality in an efficient, non-invasive way. Now, a new virtual diagnostic approach incorporates additional information about the beam that allows the method to work in situations where conventional diagnostics have failed.

U.S. Department of Energy Announces $34.5 Million for Data Science and Computation Tools to Advance Climate Solutions

The U.S. Department of Energy (DOE) today announced up to $34.5 million to harness cutting-edge research tools for new scientific discoveries, including clean energy and climate solutions. Two new funding opportunities will support researchers using data science and computation-based methods—including artificial intelligence and machine learning—to tackle basic science challenges, advance clean energy technologies, improve energy efficiency, and predict extreme weather and climate patterns.

Virtual Argonne workshop provides guidance on using AI and supercomputing tools for science

The Argonne Leadership Computing Facility continues its efforts to build a community of scientists who can employ AI and data-intensive analysis at a scale that requires DOE supercomputers.

NUS researchers harness AI to identify cancer cells by their acidity

Healthy and cancer cells can look similar under a microscope. One way of differentiating them is by examining the level of acidity, or pH level, inside the cells. Tapping on this distinguishing characteristic, a research team from the National University of Singapore (NUS) has developed a technique that uses artificial intelligence (AI) to determine whether a single cell is healthy or cancerous by analysing its pH. Each cancer test can be completed in under 35 minutes, and single cells can be classified with an accuracy rate of more than 95 per cent.