Showing 20 articles starting at article 61
< Previous 20 articles Next 20 articles >
Categories: Engineering: Robotics Research, Mathematics: Modeling
Published Study reveals why AI models that analyze medical images can be biased
(via sciencedaily.com) Original source Researchers have found that artificial intelligence models that are most accurate at predicting race and gender from X-ray images also show the biggest 'fairness gaps' -- that is, discrepancies in their ability to accurately diagnose images of people of different races or genders.
Published The future of metals research with artificial intelligence
(via sciencedaily.com) Original source A research team has developed an optimal artificial intelligence model to predict the yield strength of various metals, effectively addressing traditional cost and time limitations.
Published Light-controlled artificial maple seeds could monitor the environment even in hard-to-reach locations
(via sciencedaily.com) Original source Researchers have developed a tiny robot replicating the aerial dance of falling maple seeds. In the future, this robot could be used for real-time environmental monitoring or delivery of small samples even in inaccessible terrain such as deserts, mountains or cliffs, or the open sea. This technology could be a game changer for fields such as search-and-rescue, endangered species studies, or infrastructure monitoring.
Published New deep-learning model outperforms Google AI system in predicting peptide structures
(via sciencedaily.com) Original source Researchers have developed a deep-learning model, called PepFlow, that can predict all possible shapes of peptides -- chains of amino acids that are shorter than proteins, but perform similar biological functions. Peptides are known to be highly flexible, taking on a wide range of folding patterns, and are thus involved in many biological processes of interest to researchers in the development of therapeutics.
Published Robots face the future
(via sciencedaily.com) Original source Researchers have found a way to bind engineered skin tissue to the complex forms of humanoid robots. This brings with it potential benefits to robotic platforms such as increased mobility, self-healing abilities, embedded sensing capabilities and an increasingly lifelike appearance. Taking inspiration from human skin ligaments, the team included special perforations in a robot face, which helped a layer of skin take hold.
Published Meet CARMEN, a robot that helps people with mild cognitive impairment
(via sciencedaily.com) Original source Meet CARMEN, short for Cognitively Assistive Robot for Motivation and Neurorehabilitation -- a small, tabletop robot designed to help people with mild cognitive impairment (MCI) learn skills to improve memory, attention, and executive functioning at home.
Published Prying open the AI black box
(via sciencedaily.com) Original source Meet SQUID, a new computational tool. Compared with other genomic AI models, SQUID is more consistent, reduces background noise, and can yield better predictions regarding critical mutations. The new system aims to bring scientists closer to their findings' true medical implications.
Published Unifying behavioral analysis through animal foundation models
(via sciencedaily.com) Original source Behavioral analysis can provide a lot of information about the health status or motivations of a living being. A new technology makes it possible for a single deep learning model to detect animal motion across many species and environments. This 'foundational model', called SuperAnimal, can be used for animal conservation, biomedicine, and neuroscience research.
Published Can AI learn like us?
(via sciencedaily.com) Original source Scientists have developed a new, more energy-efficient way for AI algorithms to process data. His model may become the basis for a new generation of AI that learns like we do. Notably, these findings may also lend support to neuroscience theories surrounding memory's role in learning.
Published Simplicity versus adaptability: Understanding the balance between habitual and goal-directed behaviors
(via sciencedaily.com) Original source Scientists have proposed a new AI method in which systems of habitual and goal-directed behaviors learn to help each other. Through computer simulations that mimicked the exploration of a maze, the method quickly adapts to changing environments and also reproduced the behavior of humans and animals after they had been accustomed to a certain environment for a long time. The study not only paves the way for the development of systems that adapt quickly and reliably in the burgeoning field of AI, but also provides clues to how we make decisions in the fields of neuroscience and psychology.
Published Custom-made molecules designed to be invisible while absorbing near-infrared light
(via sciencedaily.com) Original source Researchers used theoretical calculations assessing electron orbital symmetry to synthesize new molecule designed to be both transparent and colorless while absorbing near-infrared light. This compound demonstrates the first systematic approach to producing such materials and have applications in advanced electronics. This compound also shows semiconducting properties.
Published Breakthrough approach enables bidirectional BCI functionality
(via sciencedaily.com) Original source Brain-computer interfaces or BCIs hold immense potential for individuals with a wide range of neurological conditions, but the road to implementation is long and nuanced for both the invasive and noninvasive versions of the technology. Scientists have now successfully integrated a novel focused ultrasound stimulation to realize bidirectional BCI that both encodes and decodes brain waves using machine learning in a study with 25 human subjects. This work opens up a new avenue to significantly enhance not only the signal quality, but also, overall nonivasive BCI performance by stimulating targeted neural circuits.
Published Researchers use large language models to help robots navigate
(via sciencedaily.com) Original source A technique can plan a trajectory for a robot using only language-based inputs. While it can't outperform vision-based approaches, it could be useful in settings that lack visual data to use for training.
Published Self-assembling and disassembling swarm molecular robots via DNA molecular controller
(via sciencedaily.com) Original source Researchers have succeeded in developing a DNA-based molecular controller. Crucially, this controller enables the autonomous assembly and disassembly of molecular robots, as opposed to manually directing it.
Published Quantum data assimilation: A quantum leap in weather prediction
(via sciencedaily.com) Original source Data assimilation is an important mathematical discipline in earth sciences, particularly in numerical weather prediction (NWP). However, conventional data assimilation methods require significant computational resources. To address this, researchers developed a novel method to solve data assimilation on quantum computers, significantly reducing the computation time. The findings of the study have the potential to advance NWP systems and will inspire practical applications of quantum computers for advancing data assimilation.
Published Mobile monitoring for an airborne carcinogen in Louisiana's 'Cancer Alley'
(via sciencedaily.com) Original source Louisiana's southeastern corridor is sometimes known colloquially as 'Cancer Alley' for its high cancer incidence rates connected to industrial air pollution. Most of the region's air pollution-related health risks are attributed to ethylene oxide, a volatile compound used to make plastics and sterilize medical equipment. Researchers measured concerning levels of ethylene oxide in this area with mobile optical instruments, a technique they say could improve health risk assessments.
Published Robot radiotherapy could improve treatments for eye disease
(via sciencedaily.com) Original source Researchers have successfully used a new robot system to improve treatment for debilitating eye disease.
Published Towards a new era in flexible piezoelectric sensors for both humans and robots
(via sciencedaily.com) Original source Flexible piezoelectric sensors are essential to monitor the motions of both humans and humanoid robots. However, existing designs are either are costly or have limited sensitivity. In a recent study, researchers tackled these issues by developing a novel piezoelectric composite material made from electrospun polyvinylidene fluoride nanofibers combined with dopamine. Sensors made from this material showed significant performance and stability improvements at a low cost, promising advancements in medicine, healthcare, and robotics.
Published AI-powered simulation training improves human performance in robotic exoskeletons
(via sciencedaily.com) Original source Researchers have demonstrated a new method that leverages artificial intelligence (AI) and computer simulations to train robotic exoskeletons to autonomously help users save energy while walking, running and climbing stairs.
Published 3D-printed mini-actuators can move small soft robots, lock them into new shapes
(via sciencedaily.com) Original source Researchers have demonstrated miniature soft hydraulic actuators that can be used to control the deformation and motion of soft robots that are less than a millimeter thick. The researchers have also demonstrated that this technique works with shape memory materials, allowing users to repeatedly lock the soft robots into a desired shape and return to the original shape as needed.