AI for real-time, patient-focused insight
Researchers developed a new AI model, BiomedGPT, to support various biomedical tasks. The model achieved 16 state-of-the-art results in 9 biomedical tasks and demonstrated robust predictive abilities.
Artificial Intelligence
Articles tagged with Artificial Intelligence
UW–Madison researchers find persistent problems with AI-assisted genomic studies
Researchers found that AI tools can link genetic variations with diseases like diabetes without considering the complexity of genetics. A new statistical method can reduce false positives, but proxy information studies also introduce misleading correlations, warns Qiongshi Lu.
Nanoscale transistors could enable more efficient electronics
MIT researchers develop 3D transistors using quantum mechanical properties to achieve low-voltage operation and high performance. The devices can deliver comparable performance to state-of-the-art silicon transistors while operating efficiently at much lower voltages.
AI-powered system detects toxic gases with speed and precision
The AI-powered system can detect toxic gases like nitrogen dioxide in real-time, identifying the source of harmful gas leaks. The system's optimization technique ensures fewer resources are used while providing faster and more accurate gas leak detection.
LiU researchers make AlphaFold predict very large proteins
Researchers at Linköping University have developed a new version of AlphaFold that can predict the shape of very large and complex protein structures, integrating experimental data. This breakthrough aims to improve the development of new proteins for medical drugs.
Security in quantum computing
Researchers at NCSA have presented a novel post-quantum cryptography network instrument to measure PQC adoption rates and ensure secure data safeguarding. The project's findings indicate that only OpenSSH and Google Chrome have successfully implemented PQC, achieving an initial adoption rate of 0.029%.
Argonne to help drive AI revolution in astronomy with new institute led by Northwestern University
The NSF-Simons AI Institute for the Sky (SkAI), led by Northwestern University, will develop innovative AI tools capable of handling vast data from astronomical surveys. Funded by a $20 million grant, Argonne National Laboratory will help drive this revolution in astronomy.
From compliance to conversation: New guidelines push for ethical reflection in research reporting
A new study calls for the adoption of new research ethics policies to foster learning and discussion of ethical issues. The guidelines aim to shift from compliance-based ethics to promoting ethical norms and practices.
Textbooks come alive with new, interactive AI tool
A team of computer scientists has developed an interactive AI tool that transforms static physics textbook diagrams into 3D simulations, allowing students to visualize complex concepts like momentum and electrical currents in action.
An AI assistant to guide researchers in labyrinthine world of computational modeling and discovery
A new generative AI tool called Theseus is being developed to reduce time and labor in computational modeling, suggesting and running simulations to accelerate research and discovery. The AI assistant aims to enhance the work of computational scientists by providing scientific intuition, checking code for bugs and making suggestions.
New methods for whale tracking and rendezvous using autonomous robots
A new reinforcement learning framework, AVATARS, uses diverse sensor streams to track whales and predict their surface locations. The framework aims to minimize missed encounters and improve data collection for Project CETI.
Longevity biotechnology: AI, biomarkers, geroscience & applications for healthy aging
Researchers synthesizing AI with biomarker analysis aim to detect early signs of aging and enable targeted interventions to delay disease. This approach has the potential to slow, prevent or reverse certain effects of aging.
Wearable ultrasound tech for muscle monitoring opens new possibilities in healthcare and human-machine interfaces
Researchers developed a wearable ultrasound device that tracks muscle function without invasive procedures, offering high-resolution imaging and wireless monitoring capabilities. The technology has potential applications in respiratory health and human-machine interfaces.
Rice, Texas Medical Center institutions jointly award seed grants
Two teams, led by Rice's ENRICH office, have won $100,000 in matching funds from partnering institutions to investigate AI-based symptom checkers and electrocardiogram data for improving cardiovascular risk assessment. The projects aim to mitigate vulnerabilities and reduce health disparities.
New Illinois study explores adoption of robotic weeding to fight superweeds
A new University of Illinois study examines the adoption of robotic weeding technology to fight superweeds, finding that forward-looking management strategies are more effective than myopic approaches. The research suggests that farmers with a long-term perspective on weed resistance will benefit from early adoption of robots.
Accounting for bias in medical data helps prevent AI from amplifying racial disparity
Researchers found that Black patients are less likely to receive medical tests, leading to inaccurate AI models. A new algorithm corrects for this bias by identifying untested patients based on race and vital signs, improving model accuracy to around 60%.
Novel artificial intelligence-based method for pathological diagnosis of hereditary kidney diseases
Researchers developed an AI-based method to analyze kidney lesions in female patients with Alport syndrome, predicting renal prognosis and guiding treatment interventions. The approach uses a modified stain and deep learning to detect basement membrane lesions, showing a positive correlation with proteinuria concentration.
Personal info and privacy control may be key to better visits with AI doctors
Researchers found that AI doctors who recall patients' social information, with privacy control offered, increase patient satisfaction. Human doctors don't require social or medical information to establish a close relationship with patients.
How virtual cows could help us improve human-robot interactions
Researchers used a virtual cow game to study human movement and navigation, developing a model that can simulate human behavior and predict choices. The study found that humans make decisions based on angular distance and previous choices, and the developed model could accurately mimic these patterns.
Smarter clot busting: WPI researcher to develop real-time imaging technology for stroke treatment
Yihao Zheng and his team are developing a fiber-optic probe that analyzes artery blockages in the brain and guides procedures for blockage removal. The technology uses light and advanced calculations to determine the properties of blood clots, enabling doctors to make informed decisions about how to remove them.
Using AI to measure prostate cancer lesions could aid diagnosis and treatment
Researchers developed an AI model that can identify and measure aggressive prostate cancer lesions with high accuracy. The model's estimates of tumor size were associated with the likelihood of cancer recurrence or metastasis.
AI assistants offer challenging choice: Japanese consumers value transparency over performance and sustainability
A study by Osaka University reveals that Japanese consumers value transparency in AI assistants, compromising on performance for greater clarity. Environmental sustainability is also a consideration, but remains secondary to cost and performance.
Humans and AI: Do they work better together or alone?
A new study from the MIT Center for Collective Intelligence found that combining humans and AI can be effective in certain tasks, such as creative ones like summarizing social media posts or generating new content. However, human-AI combinations often underperform compared to AI alone in decision-making tasks. The researchers suggest t...
AI algorithm accurately detects heart disease in dogs
A machine learning algorithm developed by University of Cambridge researchers can detect and grade heart murmurs in dogs with high accuracy, similar to expert cardiologists. The technology has the potential to empower primary care veterinarians to provide early detection and treatment, improving quality of life for dogs.
University of Houston engineers unveil AI model for predicting, controlling pandemic spread
A team of University of Houston engineers developed an AI tool to predict and control pandemic spread by analyzing international air travel. The analysis found that reducing flights in Western Europe can lead to fewer global COVID-19 cases, making it a key strategy for controlling the pandemic.
New AI model could make power grids more reliable amid rising renewable energy use
Researchers developed an AI model that addresses uncertainties in renewable energy generation and electric vehicle demand, making power grids more reliable. The model uses multi-fidelity graph neural networks to optimize solutions within seconds, improving grid performance even under unpredictable conditions.
New AI tool predicts protein-protein interaction mutations in hundreds of diseases
A new AI tool called PIONEER has been developed to predict protein-protein interaction mutations in hundreds of diseases, including dozens of cancers. The tool allows researchers to navigate the interactome for over 10,500 diseases and identify potential drug targets.
Solving computationally hard problems with 3D integrated photonics
Researchers have developed a reconfigurable three-dimensional integrated photonic processor specifically designed to tackle the subset sum problem, a classic NP-complete challenge. The processor operates by allowing photons in a light beam to explore all possible paths simultaneously, providing answers in parallel and demonstrating hig...
AI-related maternal healthcare software improves odds of good care by 69%, research finds
Using Clinical Design Support Systems (CDSS) can help make maternity services safer for pregnant women, with a meta-analysis of 35 studies showing a 1.69 times higher odds of improved outcomes. The findings support the use of AI and related software tools to improve pregnancy care.
AI eye to eye with ophthalmologists in diagnosing corneal infections, study finds
A new study finds that AI-powered models exhibit similar levels of accuracy as ophthalmologists in identifying infectious keratitis, a leading cause of corneal blindness worldwide. The AI models displayed a sensitivity and specificity of 89.2% and 93.2%, respectively, matching the diagnostic accuracy of human experts.
Artificial intelligence is creating a new way of thinking, an external thought process outside of our minds
Artificial intelligence is creating a new way of thinking with System 0, an external cognitive process that enhances human abilities. This system relies on AI's capabilities to process information and provide suggestions, but humans must interpret and assign meaning to the results.
Adapting GenAI for the next generation of learning
A new study by Monash University researchers highlights essential factors for adapting Generative AI (GenAI) to support human learning in all levels of education and workplaces. The key considerations include understanding how to use GenAI to enhance human learning while fostering critical thinking and self-reflection skills.
A fully automated AI-based system for assessing IVF embryo quality
A new AI-based system, BELA, accurately assesses IVF embryo quality by analyzing time-lapse video images and maternal age. The system generates a predictive score for euploidy or aneuploidy, offering an objective measure of embryo quality.
Listening skills bring human-like touch to robots
A new system dubbed SonicSense allows robots to interpret the world through acoustic vibrations, giving them a richer ability to 'feel' and understand objects. The system, featuring a robotic hand with contact microphones, can identify materials, shapes, and recognize objects in complex environments.
Study: AI could transform how hospitals produce quality reports
A pilot study found that AI tools can accurately process hospital quality measures, achieving 90% agreement with manual reporting. This can lead to more efficient and reliable approaches to healthcare reporting, especially for complex measures like severe sepsis and septic shock.
AI-assisted colonoscopy increases polyp and adenoma detection in routine screening
A systematic review of randomized clinical trials found that AI-assisted colonoscopies increased overall detection of colonic polyps and adenomas, with a small increase in procedure time. Researchers also found no clear differences in benefit for detecting adenomas across different AI systems.
Cloud computing captures chemistry code
A team of scientists and experts led by PNNL has developed a cloud computing approach to democratize access to emerging resources. They demonstrated that cloud computing can provide an agile complement to high-performance computing facilities, enabling complex chemistry workflows to be completed in days instead of months. The initiativ...
Preventive medicine professors part of collaborative grant for AI system to enhance Alzheimer's caregiving
A team of researchers has developed an AI-powered system to provide evidence-based caregiver training information and support to individuals with Alzheimer's and related dementia. The system is designed to be personalized and accessible, using validated resources and tools to address the unique needs of caregivers.
Penn engineering research discovers critical vulnerabilities in AI-enabled robots
Penn Engineering researchers found that certain features of AI-governed robots have security vulnerabilities, putting safety at risk. The 'jailbreak' rate of OpenAI's ChatGPT was 100% in just days, bypassing safety guardrails and allowing manipulation or hacking.
Most older adults don’t trust AI-generated health information — but many aren’t sure what to trust
A recent poll found that 74% of people over 50 would not trust health information generated by artificial intelligence. Only 32% of older adults reported it's very easy to find something accurate online. Health systems, academic institutions, and government agencies can use the results to improve health information accessibility.
UVA researchers engineer AI breakthrough in human action detection technology
Researchers at UVA's School of Engineering and Applied Science have developed an AI-driven intelligent video analyzer capable of detecting human actions with unprecedented precision and intelligence. The system, called SMAST, promises to transform industries such as surveillance, healthcare, and autonomous driving.
Project to integrate human and machine intelligence to address information integrity
A new project, 'Crowd-Assisted Human-AI Teaming with Explanations,' aims to develop an interactive AI system that leverages the collective strengths of human crowd workers and machine learning models. The researchers will use crowdsourcing platforms to recruit experts and non-experts to perform tasks, making the system more robust and ...
Training medical AI with knowledge, not shortcuts
Researchers develop Knowledge-enhanced Bottlenecks (KnoBo) method to emulate human physicians' education, resulting in more accurate and interpretable AI models for medical image recognition. KnoBo-based models outperform existing best-in-class models on accuracy and robustness, especially in handling confounded data.
FDA perspective on the regulation of AI in health care and biomedicine
The FDA will continue to play a central role in regulating AI technologies that improve health outcomes. The agency emphasizes the need for rigorous oversight of these transformative tools, crucial for protecting patients and clinicians.
Rice-led research will leverage responsible AI to enhance coastal communities’ severe storm response
Researchers at Rice University develop OpenSafe.AI, a system that leverages responsible AI and hazard models to provide timely insights for emergency response organizations and communities before, during, and after tropical cyclones and coastal storm events. The goal is to enable better preparation and navigation for severe weather.
Touching the future: Mastering physical contact with new algorithm for robots
Researchers at the University of Pennsylvania have developed a new consensus complementarity control (C3) algorithm that allows robots to react to complex physical contact in real-time. The algorithm enables robots to control the motion of sliding objects, a challenging task previously thought to be impossible for autonomous robots.
AI meets medicine
Researchers at Penn Engineering and PSOM will develop AI systems to predict treatment response in breast cancer, heart attacks, and sepsis. The goal is to support clinicians with accurate and transparent predictions, leading to better health outcomes for patients.
Non-invasive eye test for multiple diseases to advance under $4.8 million NIH award
Researchers at Indiana University are developing next-generation ophthalmoscopes to spot early warning signs of diseases like Alzheimer's, diabetes, and heart disease with a simple eye scan. The technology uses machine learning and AI to reduce diagnosis time from days to minutes.
Real-time descriptions of surroundings for people who are blind
WorldScribe, a new software, uses generative AI to provide real-time text and audio descriptions of surroundings for people who are blind or have low vision. The tool can adjust the level of detail based on user commands or camera frame time.
AI-trained CCTV in rivers can spot blockages and reduce floods
Researchers have developed AI-enabled detection software that can accurately detect natural debris, litter, or waste blocking culverts. The system can be integrated to existing CCTV systems to provide proactive flood defense, improving safety for response teams.
Having a sweet tooth is linked to higher risk of depression, diabetes, and stroke, study finds
A study from the University of Surrey found that people with a preference for sweets are at a higher risk of developing depression, diabetes, and suffering a stroke. The researchers used artificial intelligence to group 180,000 volunteers into three profiles: health-conscious, omnivore, and sweet tooth.
WPI researcher receives National Science Foundation grant to study how humans interact with artificial intelligence
The two-year study aims to explore biases in AI systems and develop a 'human-in-the-loop' framework for quality data discovery. It will investigate how humans can be involved as labelers, prompters, and validators to improve data sets and user interfaces.
Will AI tools revolutionize public health? Not if they continue following old patterns, researchers argue
Researchers argue that AI's impact on public health depends on its development and deployment, warning of potential biases, lack of transparency, and exacerbation of existing harm. However, the authors also highlight AI's potential to improve healthcare systems, access to care, and provide essential services.
Yao developing language model-powered virtual mathematics classroom
Ziyu Yao is creating a language model-powered virtual mathematics classroom to support collaborative learning and promote more equitable education for middle school students from under-resourced communities. The platform aims to provide effective mathematics education and explore the opportunities and risks of Generative AI techniques.
Latest advances in brain network models for medical applications: A comprehensive review highlights future potential
Recent advances in Brain Network Models (BNMs) have improved simulations of brain activities, understanding neuropathological mechanisms, and predicting disease progression. BNMs integrate structural and functional connectivity data to analyze abnormal network dynamics.
UW–Madison researchers use AI to identify sex-specific risks associated with brain tumors
Researchers have used AI to analyze digital images of pathology slides and identified risk factors for aggressive brain tumors that differ between men and women. These findings could lead to more individualized care and improved treatment outcomes for glioblastoma patients.
AI and quantum mechanics team up to accelerate drug discovery
Researchers created SmartCADD, an AI-powered virtual tool combining quantum mechanics and Computer Assisted Drug Design techniques. The tool speeds up the screening of chemical compounds, significantly reducing drug discovery timelines and identifying promising HIV drug candidates.
George Mason researchers conducting AI exploration for snow water equivalent
Researchers at George Mason University are developing an AI-powered system to predict snow water equivalent (SWE) forecasts, utilizing graph neural network-based models and physics-based constraints. The project aims to create more accurate and reliable SWE forecasts by capturing detailed snow accumulation and melt processes.
Revolutionizing cardiovascular risk assessment with AI
A recent study explores the potential of AI in ophthalmology to detect elevated HbA1c levels, a marker for high blood sugar and cardiovascular disease. The research highlights the importance of developing trustworthy AI models and addressing challenges in clinical adoption.
Unleashing the power of generative AI on smart collaborative innovation network platform to empower research and technology innovation
The CityUHK team develops a new federated learning technology integrated with research knowledge graphs and large language models to digitally transform STI services. This will create a comprehensive, domain-specific knowledge base and provide intelligent assistant services for researchers.