Add BrightSurf on Google Email

How AI support can go wrong in safety-critical settings

A new study suggests that adopting AI in high-stakes settings like hospitals and airplanes requires evaluating algorithms and human decision-making simultaneously. The study found that accurate AI predictions improved participant performance by 50-60%, but inaccurate predictions led to a 100% degradation in proper decision making.

SourceOhio State University·Journalnpj Digital Medicine·DateAug 18, 2025

Automatic design of metaheuristics: The future of optimization?

A review published in Intelligent Computing outlines the strengths of automatic approaches to designing metaheuristics, which can lead to more successful outcomes and reduce redundant, metaphor-based algorithms. The authors encourage research that relies on automatic design, utilizing modular software frameworks and configuration tools.

SourceIntelligent Computing·JournalIntelligent Computing·TypeLiterature review·DateMar 14, 2024

Multi-objective multigraph feature extraction for the shortest path cost prediction

Researchers develop novel feature extraction methods for multigraphs to predict shortest path costs, improving airport operations and sustainability. The proposed statistics-based and learning-based approaches show promising results, outperforming traditional exact search algorithms in computational efficiency.

SourceGreen Energy and Intelligent Transportation·JournalGreen Energy and Intelligent Transportation·TypeData/statistical analysis·DateMar 12, 2024

Anastasios Kyrillidis wins NSF CAREER Award

Anastasios Kyrillidis has won a National Science Foundation CAREER Award to explore the theory and design of non-convex optimization algorithms. His research aims to devise algorithmic foundations and theory that will accelerate problem-solving in machine learning, information processing, and optimization.

Giving AI penalties to get better diagnoses

A new study improves AI diagnoses by penalizing algorithms for false negatives, which can be more urgent than accuracy. Researchers achieved significant improvements in precision and recall for chronic kidney disease and other conditions using cost sensitivity techniques.

SourceUniversity of Johannesburg·JournalInformatics in Medicine Unlocked·TypeData/statistical analysis·DateNov 1, 2021

Artificial intelligence-based technology quickly identifies genetic causes of serious disease

A new AI-powered algorithm, GEM, has been developed to quickly identify genetic causes of serious disease in newborns. The technology leverages machine learning and natural language processing to analyze vast amounts of genomic data and clinical records, achieving an accuracy rate of 92% compared to existing tools.

SourceUniversity of Utah Health·JournalGenomic Medicine·TypeData/statistical analysis·DateOct 13, 2021