Articles tagged with Artificial Intelligence
Researchers at Radboud University Medical Center found that AI detects breast cancer on mammograms more frequently than radiologists. The study analyzed 42,000 breast scans and showed that one radiologist working with AI detects more tumors than two radiologists alone.
Experts from across industries and disciplines gathered to assess the current state of cybersecurity through a multidisciplinary lens. The workshop explored emerging technologies and the intersection of technology and policy, highlighting the need for a unified ecosystem to address persistent challenges.
The study reveals that East and West coasts, Great Lakes and Gulf regions face the highest risk of weather-induced blackouts. The nationwide Power System Vulnerability Index (PSVI) tool identifies areas at increased risk of power outages, enabling policymakers to prioritize preparation for long and frequent power outages.
Researchers designed over 36 million compounds using generative AI algorithms and computationally screened them for antimicrobial properties. The top candidates show promise against Neisseria gonorrhoeae and Staphylococcus aureus (MRSA) by novel mechanisms that disrupt bacterial cell membranes.
Researchers developed a novel approach called R3DG that analyzes representations at varying granularities to capture nuanced emotional fluctuations and reduce computational complexity. This framework demonstrates superior performance in multiple multimodal tasks, including sentiment analysis, emotion recognition, and humor detection.
A team at Osaka Metropolitan University used AI to analyze video footage of calves, detecting consistent temperature changes in their eyes and noses. The method enables non-invasive monitoring and improves the accuracy of body temperature assessments.
Researchers have developed a new AI approach called HEAT-ML that accelerates calculations of magnetic shadows in fusion vessels, enabling faster design and operation. This breakthrough could lead to significant improvements in fusion power generation and potentially limitless clean energy.
Researchers at King's College London found that AI chatbots can be exploited to reveal up to 12.5 times more personal information from users. The study used three types of malicious AIs to encourage disclosure of private info, with reciprocal strategies proving the most effective.
Researchers leverage machine learning to accelerate solid-state battery development, overcoming material interactions and interface instability. AI-powered strategies optimize battery management systems and decode ion transport for improved energy storage.
Researchers have developed an AI system that can recognize the early warning stages of laryngeal cancer from voice recordings. By analyzing variations in tone, pitch, and volume, the AI can distinguish between voices with benign vocal fold lesions and those with cancer. The study's findings offer a promising breakthrough for non-invasi...
Researchers at the University of Pennsylvania used AI to identify previously unknown compounds in Archaea that could fuel the development of next-generation antibiotics. The study, published in Nature Microbiology, found that 93% of the identified archaeasins demonstrated antimicrobial activity against drug-resistant bacteria.
A research team developed a new method to precisely edit DNA by combining genetic engineering with artificial intelligence. The technique enables accurate modeling of human diseases and lays the groundwork for next-generation gene therapies.
Researchers evaluated over 1,500 AI algorithms for detecting breast cancers on mammography images, achieving median rates of 98.7% specificity and 27.6% sensitivity. Combining top-performing algorithms resulted in sensitivity boosts to 60.7% and 67.8%, comparable to average screening radiologists.
Catal-GPT streamlines catalyst formulation generation and optimizes data-driven industrial catalysis. The AI-powered platform achieves high accuracy in knowledge extraction from scientific literature and generates experimental protocols.
Researchers from FAU and the Sensing Institute created a detailed computer model of the mouse brain's vasculature, simulating how brain blood vessels respond to hemodynamics and vasodynamics. The model shows that transitional vessels play a critical role in regulating flow and protecting the brain during increased activity.
A study published in The Lancet Gastroenterology & Hepatology found that routine AI assistance during colonoscopies can lead to a decrease in the ability of endoscopists to detect precancerous growths without AI aid. This reduction in skills may impact patient-related outcomes and highlight the need for further research on the long-ter...
A study of 1,100 professionals reveals that AI tools can make managers' emails more professional but also undermine trust when used for routine communication tasks. Employees perceive AI-generated content as less sincere and question their leadership abilities.
The Intelligent Data Exploring Assistant (IDEA) framework combines large language models with scientific data to analyze complex geoscience data. Researchers can ask IDEA to retrieve data, run analyses, and generate plots using plain-language questions.
A study by University of Arizona researchers found that ChatGPT 3.5 was less likely to correctly answer application-based questions (44%) compared to recall of facts (80%). Pharmacy students outperformed ChatGPT in answering case-based questions correctly.
Researchers are combining machine learning algorithms with neuromorphic hardware to build brain-like devices that can learn from data and adapt in real-time. These devices have the potential to revolutionize industries such as manufacturing by enabling machines to sense their environment, adapt to new tasks, and make decisions without ...
Scientists have developed a system that uses gene editing to create flowers that can be easily pollinated by AI-controlled robots, addressing a long-standing pollination hurdle. This breakthrough reduces manual labor costs in crops like tomatoes and soybeans, and has the potential to increase crop yields and improve sustainability.
Researchers synthesized AI-driven analysis of protein structures to predict drug interactions and optimize synergy. This fusion approach aims to solve drug resistance, reduce side effects, and guide personalized treatment in precision medicine.
Large language models are revolutionizing bioinformatics research by decoding complex biological data, predicting protein structures, and identifying disease-linked genes. The models' five core strengths include enabling breakthroughs in protein analysis, drug discovery, and genomic research.
Researchers from Shibaura Institute of Technology used machine learning algorithms to predict bearing layer depth and assess liquefaction risk. The study found that random forest models outperformed others, especially with increasing spatial data density.
Researchers found that large language models like ChatGPT outperform humans in identifying neuromyths, but struggle to critically correct them. By prompting AI to explicitly correct false assumptions, error rates can be significantly reduced.
Researchers developed a machine learning model to predict liquid crystalline polyimides with high thermal conductivity, achieving 96% accuracy. The model identified six promising candidates, which demonstrated up to 1.26 W/mK thermal conductivities, accelerating the development of efficient thermal materials.
Researchers are developing an AI imaging approach to distinguish between tumor progression and treatment effects in glioblastoma patients, promising to improve care and outcomes. The AI approach has already shown accuracy rates of up to 74% in initial testing.
A new study combines eye tracking and artificial intelligence to identify key moments in educational videos that matter for learning. The research provides promise for individualized video education, where AI dynamically adjusts the content for a student's understanding.
Researchers developed a new speculative decoding framework called SPECTRA, which optimizes internal and external speculation for accelerated text generation. By integrating a core module and an optional retrieval module, SPECTRA achieves substantial speedups across diverse tasks and model architectures while preserving output quality.
A large-scale modeling study led by MIT researchers reveals that dynamically adjusting vehicle speeds can cut annual city-wide intersection carbon emissions by 11-22%. Implementing eco-driving measures could also result in a 25-50% reduction in CO2 emissions if only 10% of vehicles adopt the technology.
A new principle for AI regulation, multiplicity, aims to design systems that expose users to a broader range of narratives, support diverse alternatives, and encourage critical engagement. This can help enrich the human experience and prevent the suppression of diverse perspectives, cultural diversity, and social tolerance.
Researchers developed a deep learning model to predict ADRs using molecular structures, achieving high accuracy in detecting rare effects and minimal risk for novel molecules. The model demonstrated potential as a decision-support tool in early-phase drug discovery and regulatory safety monitoring.
Researchers found that AI-augmented, personalized feedback from human instructors enhanced surgical performance and risk management. This study suggests the integration of human expertise in AI-driven education can improve engagement and ensure contextualized, adaptive feedback.
A University of Washington study found that biased AI chatbots can influence users' political views, with those who interacted with the most biased systems shifting their views more significantly. The researchers suggest education about AI may help mitigate this effect.
A study by researchers at UC San Diego and University of Chicago found that most visual artists lack the technical know-how to block AI crawlers, despite wanting to protect their creative work. The team developed a tool called Glaze, but more effective solutions are needed to prevent AI data harvesting.
A new deep learning model, MSI-SEER, achieves high accuracy in predicting microsatellite instability-high tumors and their responsiveness to immunotherapy. The model combines tumor MSI status with stroma-to-tumor ratio for highly accurate ICI response prediction.
The system tracks and analyzes crop development using data from sensors, biosensors, the Internet of Things, and AI. Strong security protocols ensure farmer data remains private and resilient against future quantum computer attacks. The research team plans to improve their system with faster sensor processing and a solar-powered battery.
The Ateneo de Manila University's Business Insights Laboratory explores how AI can turn handwritten sales logs into manageable digital data. The system uses OCR and LLM technology to recognize products, match prices, and tabulate sales summaries, helping businesses quickly identify bestsellers or slow-moving stock.
Researchers developed a novel framework using generative AI and musculoskeletal simulation to create synthetic gait data. This approach enables more robust and generalizable gait analysis across various patient populations and clinical environments.
Researchers have developed a novel synthetic ligand using AI-based computational protein design, activating the Notch signaling pathway in T-cells. This breakthrough enables broadened clinical T-cell production and advances immunotherapy development.
Researchers at NJIT used artificial intelligence to discover new porous materials capable of revolutionizing multivalent-ion batteries. The AI-driven approach uncovered five entirely new materials with large, open channels ideal for moving bulky multivalent ions quickly and safely. These findings offer a promising solution for the futu...
Physicists used a machine-learning method to identify surprising new twists on the non-reciprocal forces governing a many-body system. The AI approach provides precise approximations for these forces, correcting common theoretical assumptions with an accuracy of over 99%.
The article provides reporting recommendations for studies evaluating chatbot health advice, highlighting the importance of standardization in interpretability. Key findings include the need for clear guidance on chatbot performance metrics and data quality control.
Hollings researchers use natural language processing to identify primary cancer types in medical notes, improving treatment accuracy and personalized care. The AI model achieves high accuracy rates, outperforming standard medical codes in identifying primary cancer diagnoses.
A new AI tool can learn to read medical images with far less data, cutting down the amount of required data by up to 20 times. The tool improves upon medical image segmentation, a labor-intensive task often performed by experts, and boosts model performance in settings with limited annotated data.
The survey highlights a dual-path framework for harnessing large language models' intrinsic reasoning and integrating them with external methodologies. Large language models are employed for step-by-step reasoning methods, enabling the decomposition of complex tasks and exploration of multiple reasoning paths.
A technology expert predicts AI-based medicine will revolutionise care for Alzheimer’s and diabetes, but access must be made available to all patients. The author highlights key innovations such as diagnostic imaging and surgical robots that will aid in more accurate diagnoses and treatments.
The ERIC system combines doorbell cameras and AI to analyze rainfall estimation and automatically adjusts irrigation schedules for more precise water use. Researchers estimate users can save up to $29/month in utility costs and 9,000 gallons of water per month with the innovative irrigation system.
Researchers used ChatGPT to assess patient eligibility for clinical trials, identifying suitable candidates within minutes. The AI was more accurate than GPT-3.5 but slightly slower and more expensive.
A research team developed an innovative unsupervised model for industrial anomaly detection using paired well-lit and low-light images. The model leverages feature maps, Low-pass Feature Enhancement, and Illumination-aware Feature Enhancement to detect anomalies while remaining lightweight and memory-efficient.
A new study by MIT researchers introduces a method for machine learning with symmetry that is provably efficient, clarifying a foundational question in the field. This approach could lead to more powerful AI models designed to handle symmetry, benefiting applications such as drug discovery and materials science.
Researchers at Salk Institute launched a machine learning framework called ShortStop to explore overlooked DNA regions and discover microproteins with potential roles in disease. The tool identified 210 new microprotein candidates in lung cancer data, including one validated target for therapeutic treatment.
Researchers argue that successful AI policy requires scientific understanding and systematic analysis to inform decision-making. The authors propose mechanisms to incentivize rigorous model evaluations, increase transparency, and monitor post-deployment harms to achieve evidence-based governance.
Researchers at Cornell University have developed a way to watermark light in videos, allowing fact-checkers to detect if video footage is fake or has been manipulated. The technique involves hiding information in nearly-invisible fluctuations of lighting at important events and locations, which can be used to reveal any malicious editing.
Recent AI-based models, such as Asian Prostate Cancer Artificial Intelligence and Galen Prostate, optimize screening and reduce unnecessary biopsies. These models use multimodal clinical parameters and convolutional neural networks to detect prostate cancer and identify aggressiveness of cancer cells.
The journal is seeking high-quality submissions that explore the potential of voice and speech analysis in diverse medical applications. Submissions focus on acoustic and voice analysis techniques, voice or vocal phenotyping, and vocal biomarkers in health and medicine.
The FDA has authorized artificial intelligence devices for Alzheimer's and related dementias despite limited demographic data. Disease status, age, sex, and race were reported for fewer than half of devices, raising concerns about real-world generalizability.
The new book highlights the transformative role of artificial intelligence (AI) and machine learning (ML) across various domains, including mechatronics, cybersecurity, digital health, and automation. Readers will gain practical insights into AI-based techniques in power systems, social media management, and healthcare diagnostics.
A new study reveals that large language models can generate convincing but biased peer reviews, nearly indistinguishable from human writing. The AI excelled at generating persuasive rejection comments and fabricating reasons to cite unrelated studies, posing a serious risk to the integrity of scientific publishing.
Gemini achieved highest accuracy, followed by Claude-3-opus and GPT, with accuracy varying significantly across single-image tasks and continuous slices. Simplified prompts improved diagnostic performance, but Gemini and GPT struggled with normal image recognition.