English
Arabic
Chinese (Simplified)
Dutch
Esperanto
French
German
Greek
Italian
Japanese
Korean
Portuguese
Russian
Spanish
UrduAs the year comes to a close, the U.S. Department of Energy’s Argonne National Laboratory reviews some of its most notable achievements of 2024.
Tag: Deep Learning
Breaking Barriers: Study Uses AI to Interpret American Sign Language in Real-time
A study is the first-of-its-kind to recognize American Sign Language (ASL) alphabet gestures using computer vision. Researchers developed a custom dataset of 29,820 static images of ASL hand gestures. Each image was annotated with 21 key landmarks on the hand, providing detailed spatial information about its structure and position. Combining MediaPipe and YOLOv8, a deep learning method they trained, with fine-tuning hyperparameters for the best accuracy, represents a groundbreaking and innovative approach that hasn’t been explored in previous research.
The best AI strategy to recognize multiple objects in one image
Image classification is one of AI’s most common tasks, where a system is required to recognize an object from a given image. Yet real life requires us to recognize not a single standalone object but rather multiple objects appearing together in a given image.
This reality raises the question: what is the best strategy to tackle multi-object classification? The common approach is to detect each object individually and then classify them. But new research challenges this customary approach to multi-object classification tasks.
In an article published today in Physica A, researchers from Bar-Ilan University in Israel show how classifying objects together, through a process known as Multi-Label Classification (MLC), can surpass the common detection-based classification.
Scientists Create Model to Make MRI More Accurate, Reliable
The new model, developed by researchers at the UNC School of Medicine, can produce more accurate and reliable analysis of brain structures, which is critical for early detection, medical diagnosis, and neurological research.
New transformer-based AI model enhances precision in rice leaf disease detection
A research team has developed an innovative AI model called AISOA-SSformer that significantly improves the accuracy of detecting rice leaf diseases.
Revolutionary 3D leaf edge reconstruction method enhances plant morphology analysis
A research team has developed a novel method for reconstructing the 3D edges of leaves using advanced deep learning and image processing techniques.
AI-powered model revolutionizes rice lodging detection for improved agricultural outcomes
A research team has introduced a cutting-edge deep learning model, AAUConvNeXt, for precise and efficient detection of rice lodging, a critical agricultural challenge that impacts yield prediction and disaster management.
Advanced AI techniques enhance crop leaf disease detection in tropical agriculture
A research team showcases the application of deep learning models in identifying leaf diseases in key tropical crops such as coconut, mango, and durian, offering crucial insights for the future of precision agriculture.
Are brain delays a computational disadvantage?
Biological components are less reliable than electrical ones, and rather than instantaneously receive the incoming signals, the signals arrive with a variety of delays.
Research to use machine learning to ’reverse-engineer’ new composite materials
Professors at Binghamton University, State University of New York have received NSF grant for deep-learning model that can customize microarchitecture based on specific needs
Feet First: AI Reveals How Infants Connect with Their World
Researchers explored how infants act purposefully by attaching a colorful mobile to their foot and tracking movements with a Vicon 3D motion capture system. The study tested AI’s ability to detect changes in infant movement patterns. Findings showed that AI techniques, especially the deep learning model 2D-CapsNet, effectively classified different stages of behavior. Notably, foot movements varied significantly. Looking at how AI classification accuracy changes for each baby gives researchers a new way to understand when and how they start to engage with the world.
New imaging technique brings us closer to simplified, low-cost agricultural quality assessment
A team of University of Illinois Urbana-Champaign researchers has developed a method to reconstruct hyperspectral images from standard RGB images using deep machine learning. This technique can greatly simplify the analytical process and potentially revolutionize product assessment in the agricultural industry.
Beyond Conventional Pathology, Label-free Histology Meets AI
POSTECH team develops deep-learning powered label-free photoacoustic histology for virtual staining, segmentation, and classification of human liver cancers.
$4.9-Million NSF Award Funds Major Enhancement to Bridges-2 System
$4.9 million from the NSF has funded an upgrade to PSC’s flagship Bridges-2 supercomputer. The grant has allowed the center to add late-model powerful NVIDIA H100 GPUs to the system, further enhancing its ability to support research in and requiring artificial intelligence, particularly in the context of massive data and high-performance computing.
Monitoring of nature reserves via social media and deep learning
Researchers from the National University of Singapore have created a deep learning method to analyse social media images taken within protected green spaces to gain insights on human activity distribution as a way to monitor the ecological impacts of these activities.
Argonne’s AI Testbed gives researchers access to cutting-edge AI systems for science
The Argonne Leadership Computing Facility’s AI Testbed is a growing collection of some of the world’s most advanced AI accelerators available for open science.
Advanced DeepLabv3+ Algorithm Enhances Safflower Filament Harvesting with High Accuracy
A research team has developed an improved DeepLabv3+ algorithm for accurately detecting and localizing safflower filament picking points.
Deep Learning Model Overcomes the Challenge of Real-World Measurements of Isotope Production Target Cooling Systems
Isotope production facilities depend on cooling for proper function of target systems during irradiation. Examining these systems is challenging due to high radiation levels during target irradiation that make real-world measurements impossible.
Media Cybernetics Unveils Next-Generation Image-Pro AI Software for Microscope-Based Research and Inspection
Media Cybernetics proudly announces the release of Image-Pro® AI, a groundbreaking image analysis software designed to enhance efficiency and accuracy in scientific research and quality inspection.
How can AI cope with changing categories?
Bar-Ilan University researchers have uncovered a new universal law detailing how artificial neural networks handle an increasing number of categories for identification. This law demonstrates how the identification error rate of such networks increases with the number of required recognizable objects.
Largest-ever antibiotic discovery effort uses AI to uncover potential cures in microbial dark matter
Almost a century ago, the discovery of antibiotics like penicillin revolutionized medicine by harnessing the natural bacteria-killing abilities of microbes. Today, a new study co-led by researchers at the Perelman School of Medicine at the University of Pennsylvania suggests that natural-product antibiotic discovery is about to accelerate into a new era, powered by artificial intelligence (AI).
Researchers use machine learning to detect defects in manufacturing
The algorithm was able to correctly identify hundreds of defects in real physical parts that have not previously been seen by the deep learning model.
AI goes underwater: transforming coral reef conservation with cutting-edge image analysis
In an era where coral reef ecosystems worldwide are under significant threat, the ability to accurately monitor and assess their health is more crucial than ever. This latest research introduces sophisticated deep learning models to enhance the precision and speed of coral reef imaging analyses, paving the way for more effective conservation strategies.
Ultrasound, ultraprecise: Advancing super-resolution imaging with deep learning
Researchers at the Beckman Institute for Advanced Science and Technology developed a new technique to make ultrasound localization microscopy, an emerging diagnostic tool used for high-resolution microvascular imaging, more practical for clinical settings. Their method uses deep learning to advance…
Argonne training program alumni find success in extreme-scale computing
Past attendees of the annual Argonne Training Program on Extreme-Scale Computing are thriving in careers across the field of high performance computing.
American nuclear power plants are among the most secure in the world — what if they could be less expensive, too?
Argonne collaborates with Purdue University on new research aimed at lowering the cost of developing small nuclear reactors.
FAU Engineering Study Employs Deep Learning to Explain Extreme Events
At the core of uncovering extreme events such as floods is the physics of fluids – specifically turbulent flows.
Can deep learning help us save mangrove forests?
Mangrove forests are an essential component of the coastal zones in tropical and subtropical areas, providing a wide range of goods and ecosystem services that play a vital role in ecology. They are also threatened, disappearing, and degraded across the globe.
Bright lights, big data: how Argonne is bringing supercomputing and X-rays together for scientific breakthroughs
Argonne’s newest supercomputer, Polaris, is up and running, and scientists using the Advanced Photon Source are already seeing faster data analysis. While the combination is paying dividends now, it points toward an upgraded APS and an even better supercomputer called Aurora.
Journal of Medical Internet Research | Can Artificial Intelligence Be Used to Diagnose Influenza?
JMIR Publications published “Examining the Use of an Artificial Intelligence Model to Diagnose Influenza: Development and Validation Study” in the Journal of Medical Internet Research, which reported that it may be possible to diagnose influenza infection by applying deep learning to pharyngeal images given that influenza primarily infects the upper respiratory system.
Argonne training program introduces AI for science to a new crowd
The Intro to AI-Driven Science on Supercomputers training series gives students hands-on experience using the Lab’s high performance computing resources.
ARVO Foundation Names 2023 Winners of Dr. David L. Epstein Award
The Association for Research in Vision and Ophthalmology (ARVO) announced today the 2023 recipients of the Dr. David L. Epstein Award:Since 2016, the Dr. David L. Epstein Award has been given annually to a well-established senior investigator with a documented history of conducting eye and vision research in glaucoma and of mentoring clinician-scientists to independent academic and research careers.
Automated epilepsy lesion detection on MRI: The MELD Project
In this episode of Sharp Waves, the ILAE podcast, Dr. Maryam Nabavi Nouri talks with Dr. Konrad Wagstyl about the MELD Project, an open-science consortium using deep learning principles to develop automated lesion detection of clinical MRI data.
CHOP and NJIT Researchers Develop New Tool for Studying Multiple Characteristics of a Single Cell
Researchers from Children’s Hospital of Philadelphia (CHOP) and New Jersey Institute of Technology (NJIT) developed new software that integrates a variety of information from a single cell, allowing researchers to see how one change in a cell can lead to several others and providing important clues for pinpointing the exact causes of genetic-based diseases.
AI and Cancer: Study Highlights Automated System to Calculate Metabolic Tumor Volume
AI-based approach could make it easier to incorporate metabolic tumor volume into clinical trials and possibly patient care
Deep learning underlies geographic dataset used in hurricane response
Over the past seven years, researchers in ORNL’s Geospatial Science and Human Security Division have mapped and characterized all structures within the United States and its territories to aid FEMA in its response to disasters.
Q&A With Vascular Surgeon Elizabeth Chou, MD
After 11 years spent in medical school, residency and fellowships, Elizabeth Chou, MD, a vascular surgeon who recently joined the Smidt Heart Institute at Cedars-Sinai, has earned her dream career. And she has no plans of stopping there. She’s on a path toward ensuring women in vascular surgery are represented—as incoming physicians and as patients.
Mount Sinai Researchers Use Artificial Intelligence to Uncover the Cellular Origins of Alzheimer’s Disease and Other Cognitive Disorders
Deep learning models represent “an entirely new paradigm for studying dementia”
Researchers combine data science and machine learning techniques to improve traditional MRI image reconstruction
University of Minnesota Twin Cities researchers have found a way to improve the performance of traditional Magnetic Resonance Imaging (MRI) reconstruction techniques, allowing for faster MRIs without relying on the use of newer deep learning methods.
Rensselaer Researchers to Address Big Data Challenges
Dr. Yangyang Xu, assistant professor of mathematical sciences at Rensselaer Polytechnic Institute, has received a $250,000 grant from the National Science Foundation (NSF) to research challenges associated with distributed big data in machine learning.Machine learning algorithms allow computers to make decisions, predictions, and recommendations on the basis of input training data without being explicitly told what information to look for in the data.
ClearBuds: First wireless earbuds that clear up calls using deep learning
University of Washington researchers created ClearBuds, earbuds that enhance the speaker’s voice and reduce background noise.
KyotoU PEGS away at catching quakes at light speed
A novel AI-based approach to detect earthquakes early uses prompt elasto-gravity signals, or PEGS, gravitational changes generated by large-mass motion in megaquakes. PEGS carry information about an ongoing earthquake at the speed of light, arriving much faster than even the fastest seismic waves.
DeepSqueak Tool Identifies Marine Mammal Calls #ASA182
As the size and number of acoustic datasets increase, accurately and quickly matching the bioacoustics signals to their corresponding sources becomes more challenging and important. This is especially difficult in noisy, natural acoustic environments. At the 182nd ASA Meeting, Elizabeth Ferguson, from Ocean Science Analytics, will describe how DeepSqueak, a deep learning tool, can classify underwater acoustic signals. It uses deep neural network image recognition and classification methods to determine the important features within spectrograms, then match those features to specific sources.
AF2Complex: Researchers Leverage Deep Learning to Predict Physical Interactions of Protein Complexes
Proteins are the molecular machinery that makes life possible, and researchers have long been interested in a key trait of protein function: their three-dimensional structure. A new study by Georgia Tech and Oak Ridge National Laboratory details a computational tool able to predict the structure protein complexes – and lends new insights into the biomolecular mechanisms of their function.
Novel Tag Provides First Detailed Look into Goliath Grouper Behavior
A study is the first to reveal detailed behavior of massive goliath groupers. Until now, no studies have documented their fine-scale behavior. What is known about them has been learned from divers, underwater video footage, and observing them in captivity. Using a multi-sensor tag with a three axis accelerometer, gyroscope and magnetometer as well as a temperature, pressure and light sensor, a video camera and a hydrophone, researchers show how this species navigates through complex artificial reef environments, maintain themselves in high current areas, and how much time they spend in different cracks and crevices – none of which would be possible without the tag.
Argonne collaborations bring computational tools to the forefront of COVID-19 research
Argonne, industry and academia collaborate to bring innovative AI and simulation tools to the COVID-19 battlefront.
Novel Model Predicts COVID-19 Outbreak Two Weeks Ahead of Time
People’s social behavior, reflected in their mobility data, is providing scientists with a way to forecast the spread of COVID-19 nationwide at the county level. Researchers have developed the first data-driven deep learning model with the potential to predict an outbreak in COVID-19 cases two weeks in advance. Feeding the mobility data to epidemiological forecasting models helps to estimate COVID-19 growth as well as evaluating the effects of government policies such as mandating masks on the spread of COVID-19.
NSF makes $20 Million investment in Optimization-focused AI Research Institute led by UC San Diego
The National Science Foundation (NSF) announced today an investment of $220 million to establish 11 artificial intelligence (AI) institutes, each receiving $20 million over five years. One of these, The Institute for Learning-enabled Optimization at Scale (TILOS), will be led by the University of California San Diego.
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.
Artificial intelligence models to analyze cancer images can take shortcuts that introduce bias for minority patients
New study of artificial intelligence tools that analyze tumor images shows how they can make inaccurate predictions based on the institution that submitted the image