Articles tagged with Artificial Intelligence
Large language models show promise in identifying housing instability from clinical notes, but AI interpretation can be incorrect and human review is still necessary. De-identification methods also require refinement to preserve privacy without losing important details.
Researchers have developed a new method for designing large artificial proteins with high accuracy, utilizing AI-based software Alphafold2. The approach combines accurate structure prediction with optimization techniques, allowing for the creation of proteins with tailored properties, such as precise binding and stability.
A study found that police radio transmissions disproportionately involve Black suspects, posing a higher privacy risk for Black people. The researchers analyzed 24 hours of human-transcribed communications and found that Black males were mentioned most often, even in majority white zones.
Insilico Medicine has received FDA clearance for ISM5939, a potential best-in-class oral small molecule inhibitor targeting ENPP1 for the treatment of solid tumors. ISM5939 demonstrated robust anti-tumor efficacy in preclinical studies and showed favorable safety profiles.
Researchers discovered that seven out of eight AI models generated significant amounts of biased text in interactions, particularly when discussing caste. The study's findings suggest a pressing need for policy regulations to ensure these models are safe for everyone, especially in sensitive tasks like hiring.
A global team, including Lehigh University researcher Xuanhong Cheng, is exploring molecular- and cellular-level changes in muscle tissue that could lead to better diagnostic tools and therapeutic options for CFS and long COVID. The team aims to develop noninvasive diagnostic tools using electrical signatures.
Researchers found that none of the four AI image generators accurately represented gender, racial, or disability diversity among real chemists. The study highlights concerns about the accuracy of AI-generated content and its potential to perpetuate stereotypes.
The Upanzi Network will fund research projects in Morocco, Botswana, and South Africa, focusing on innovation and open standard technologies. Researchers from different African regions will collaborate to develop context-specific digital solutions, addressing local needs and culture.
The Margot and Tom Pritzker Prize for AI in Science Research Excellence recognizes exceptional contributions to artificial intelligence and scientific inquiry. Awardees will receive $50,000 and be honored at the University of Chicago and Caltech Conference on AI+Science.
The study combines AI with fluorescence photography to enhance skin diagnostics, allowing for earlier detection of skin conditions such as acne, photoaging, and hyperpigmentation. AI can pinpoint early indicators of skin conditions with greater accuracy and monitor changes over time for data-driven decisions.
Researchers developed a machine-learning tool that provides accurate predictions for flood-prone areas, using historical data and weather-based predictors. The model can predict short-term river discharge with high accuracy, giving real-time data on water movement through the river.
A new paper predicts that AI will influence human evolution through natural selection, including changes in brain size, attention spans, and social behaviors. The author argues that human-AI interactions can resemble human-human social interactions, leading to potential evolutionary consequences.
Researchers use AI to predict COPD flare-ups based on daily urine tests, potentially allowing for proactive treatment and prevention. The study identifies biomarkers in urine that can forecast a flare-up seven days prior to symptoms.
A study published in Radiology found that radiologists and physicians who received local explanations from artificial intelligence (AI) systems performed better and made faster diagnoses than those who received global explanations. The researchers also found that AI advice was trusted more quickly when it provided local explanations.
Researchers have trained AI models to distinguish brain tumors from healthy tissue using convolutional neural networks and transfer learning. The models achieved an average accuracy of 85.99% at detecting brain cancer, with the ability to generate images showing specific areas in its tumor-positive or negative classification.
Researchers suggest that starting with smaller neural networks and using curriculum learning can improve performance and reduce the need for massive computing resources. This approach could lead to more resource-efficient and less energy-consuming AI systems.
The GenAI4ED project aims to revolutionize secondary education by developing a digital platform that assesses and selects GenAI-based educational software. The platform will provide personalized recommendations based on predefined criteria, taking into account both technical effectiveness and educational impact.
A new report from Curtin University highlights WA's strengths in mining and energy but warns of limitations in emerging sectors. The report calls for targeted reforms and investment to boost talent development, gender diversity and access to capital and infrastructure.
A new approach using topological data analysis (TDA) enhances the reliability and reduces bias in AI systems used for medical diagnosis in radiology. TDA captures intricate features and provides a holistic view of medical images, leading to more accurate diagnoses and equitable patient care.
Industry and academic experts discuss the potential of new materials, configurations, and integration technologies to overcome bandwidth limitations and operational robustness issues in silicon photonic modulators. These advancements are expected to impact emerging applications such as data centers, AI, quantum information processing, ...
The NIH-developed AI algorithm, TrialGPT, can identify relevant clinical trials for eligible patients and provide a clear explanation of their eligibility. The researchers found that clinicians using TrialGPT spent 40% less time screening patients while maintaining the same level of accuracy.
Ayya Azzahra of CervivAI and Neda Razavi of iSono Health have been named as the two winners of Qatar Foundation's World Innovation Summit for Health (WISH) 2024 Innovation Competitions. The awards include cash prizes and mentoring sessions with leading industry experts.
Erdogan Kaya and his team aim to create an AI curriculum for grades 4-5, focusing on emergent multilingual learners and preparing students for AI-intensive careers. The project explores the impact of the integrated curriculum on students' understanding of AI concepts and teachers' AI teaching efficacy beliefs.
New algorithm CALDERA enables leaner LLMs with improved performance, reducing costs, energy consumption and privacy risks. This allows organizations to adapt models to their specific needs without sharing sensitive data, enhancing user privacy.
Researchers have identified three distinct subtypes of Chiari type-1 malformations with distinct characteristics, enabling physicians to develop treatment protocols for patients. The study used artificial intelligence to analyze data from over 1,200 patients and identified patterns that can inform medical decision making.
The EDAI framework is a comprehensive guideline designed to embed equity, diversity, and inclusion (EDI) principles throughout the artificial intelligence (AI) lifecycle. It addresses a significant gap in current AI development and implementation practices in health and oral healthcare, ensuring AI systems are socially responsible and ...
The NJIT team's Limelight Rainforest developed a drone-delivered technology capable of identifying thousands of rainforest species in mere hours. The device uses telemetry, satellite communications, and AI to interpret data from advanced microphones, cameras, and capture systems.
The In-Space Physical AI Workshop, held at Rice University's Ion District, brought together top scientists and experts to explore AI applications in space. Key findings included the potential of AI to streamline spacecraft navigation and crew health management.
A new AI program, PanEcho, has been developed to interpret echocardiography videos independently, reducing wait times and improving timely medical care. The program demonstrated accuracy in estimating continuous echocardiographic parameters and quantifying left ventricle dimensions and function.
A new study reveals that an AI algorithm can accurately detect early-stage metabolic-associated steatotic liver disease (MASLD) in patients who meet the criteria, leaving 83% undiagnosed. This highlights the need for improved screening and diagnosis methods to prevent progression to advanced liver disease.
The NASA Goddard Lidar team has developed a system that can produce higher resolution data within a smaller space, significantly increasing efficiency compared to current models. This technology, known as CASALS, uses artificial intelligence and other technologies to allow the instrument to make its own decisions while in orbit.
Researchers found that membership inference attacks on large language models (LLMs) are not effective in measuring information exposure risks. The common method used to test LLM leaks suffers from ambiguity due to the fluidity of language, making it difficult to define a representative set of non-member candidates.
A panel of bioethicists and experts emphasize that human accountability is crucial for healthcare decisions made by AI. The importance of diverse data sets was also stressed to avoid biases in AI-enabled medical technologies. Experts stress the need for collective liability among developers, programmers, and data scientists.
Neuro-oncology experts developed guidelines to standardize AI use in brain cancer diagnosis and treatment. The recommendations aim to ensure reliable clinical trial results and protect patients.
Researchers at the University of Washington have developed a headphone prototype that can create 'sound bubbles' around the wearer, allowing them to focus on conversations without external noise. The system uses AI algorithms and machine learning to track sound sources and suppress distant noises.
The partnership aims to improve the speed and accuracy of flood damage assessments by combining multi-modal data sources and a novel vision language model. The system will enable first responders to quickly allocate resources and respond more efficiently to weather disasters.
Researchers developed a new AI-powered diagnostic system, FastGlioma, which reveals invisible cancerous tissue in brain tumor surgery. The technique may delay or prevent recurrence of high-grade tumors and improve patient survival.
Researchers developed an AI-powered model called FastGlioma that can detect residual tumor tissue with high accuracy in 10 seconds. The technology outperformed conventional methods, reducing the risk of missed tumors by nearly 75%. This innovation could change the field of neurosurgery and minimize reliance on radiographic imaging.
A new study published in JAMA Network Open found that using Chat GPT Plus does not significantly improve the accuracy of doctors' diagnoses, but it outperformed conventional methods in certain cases. The researchers suggest that physicians need more training and experience with AI to capitalize on its potential.
The joint Center for Health AI will focus on making care more personalized and effective while being more efficient and manageable for physicians. AI will be used to streamline workflows and administrative tasks, reducing burnout among healthcare workers and improving patient experience.
A study by Niigata University found that AI analysis of PET/CT images can predict the occurrence of interstitial lung disease, a serious side effect of immunotherapy. Patients with high inflammation in non-cancerous lungs are at a higher risk of developing this condition.
The study emphasizes the need for a coherent approach to understanding A.I. threats, recognizing the intricate interplay between technology and society. Experts propose involving laypeople and experts in risk assessment processes, as well as promoting social resilience to ensure better decision-making.
NeuroMechFly v2 simulates how a fruit fly navigates through its environment while reacting to sights, smells, and obstacles. The model can track moving objects visually or navigate towards an odor source, while avoiding obstacles in its path, enabling researchers to study brain-body coordination and animal intelligence.
Researchers at Tufts University have discovered that AI's poor performance in conversational skills stems from its limited exposure to transcribed spoken language. The AI is trained on vast datasets of written content but lacks the nuance of human conversation, making it struggle with timing and turn-taking.
The 12th Heidelberg Laureate Forum will bring together 200 talented young researchers from around the world to interact with renowned laureates of prestigious prizes. The event offers a unique opportunity for cross-generational scientific dialogue, networking, and collaboration.
AI models trained on limited data may misdiagnose certain populations, particularly those in underserved communities. Researchers propose six key strategies to develop fair AI models, including using diverse datasets and transparent decision-making processes.
A new system combining high-speed video and an AI algorithm can detect high blood pressure and Type 1 or Type 2 diabetes from subtle changes in blood flow, potentially leading to early detection and treatment. The technology is still in its preliminary stages but holds promise for non-invasive monitoring of health at home.
The University of Washington School of Medicine has released a flagship AI-ready dataset for a type 2 diabetes study, featuring a diverse range of participants and novel measures. The dataset includes environmental sensor data, survey responses, eye-imaging scans, and traditional biologic variables.
Researchers at Mass General Brigham developed an AI algorithm to identify long COVID cases in patients' health records, revealing a higher prevalence rate than previously thought. The tool uses 'precision phenotyping' to sift through individual records and track symptoms over time, reducing bias in diagnosis.
Researchers developed a software tool called Perseus that identifies critical tasks in AI training and slows down processors not on the path, eliminating unnecessary power use. This approach saves enough energy to power 1.1 million US homes in 2026 and reduces AI's carbon footprint.
Researchers developed an innovative automated wildlife camera network to monitor species in a Dutch coastal dunes nature reserve. The system, which combines autonomous cameras, wireless data transmission and AI-driven species identification, has been shown to be cost-efficient and efficient, providing new insights into the habitat use ...
Recent advancements in diagnostic techniques, including artificial intelligence integration, biomarker discoveries, and imaging technologies, have transformed the detection of pancreatic cancer. Key findings include improved sensitivity and accuracy of early-stage lesion identification using AI models, as well as the discovery of novel...
MIT researchers developed a theoretical foundation for methods to aggregate genes into related groups, enabling the efficient learning of underlying cause-and-effect relationships. They achieve this using only observational data, potentially paving the way for targeted treatments.
Researchers trained an AI model on goat faces in pain, achieving 62-80% accuracy. This technology may help clinicians treat pain effectively in patients who cannot speak. The goal is to develop a more generalizable system for pain assessment in animals and eventually humans.
The new AI-based Natural Language to Query (NLQ) technology in Dimensions translates user intent into a structured Boolean query, ensuring consistency and transparency. This update enables users to save time while producing accurate results and previewing their search outcomes more efficiently.
A new crowdsourcing system, FireLoc, uses a network of low-cost mobile phones to detect wildfires minutes—even seconds—after they ignite. The system prioritizes privacy and accurately maps wilderness fires to within 180 feet of their origin.
Research articles highlight the limitations of the 'tens rule' in invasion ecology, while also showcasing the growing use of crowdsourced science in environmental reviews. Additionally, a study uses artificial intelligence to identify individual animals with high accuracy, improving conservation strategies.
A recent study by Ruhr-University Bochum found that 0.052% of Twitter accounts use AI-generated profile pictures, which are often linked to fake profiles spreading disinformation and propaganda. The researchers also found that these accounts had fewer followers and were more likely to be blocked by Twitter.
Researchers developed a multimodal dataset, TimelyTale, to gather passenger-specific sensor data for context-relevant explanations. The approach effectively identified the timing and frequency of passenger demands for explanations, enabling the creation of a machine-learning model to predict the best time for providing an explanation.
Hoda Eldardiry receives $349,360 grant from NSF to develop practical competencies for students to apply ethical principles in AI system design. Her team aims to engage industry professionals to translate AI ethics into concrete decision-making.