St. Jude Children’s Research Hospital scientists developed an algorithm using machine learning to create 3D maps that reveal factors essential to gene expression in healthy and diseased cells.

To support governments facing AI challenges, an international group of experts from the University of Ottawa and Université de Montréal at IVADO analyzed key factors of AI implementation success and failure in the public sector to propose policy guidance for a transformative and resilient public administration in the age of AI and to better guard against the potentially negative effects and risks brought forth by this technology.

Bloodstream infections (BSI) can turn deadly fast, particularly for patients with weakened immune systems. A new study from Houston Methodist Research Institute finds that artificial intelligence can assist clinicians in identifying previously unseen patterns of infection in patients. 

Cobalt-free LiNiO₂ (LNO) is considered a promising cathode for its high energy density and cost-effectiveness. However, its structural instability under deep delithiation severely limits practical application in next-generation lithium-ion batteries (LIBs). Microstructure engineering enhances structural stability through precisely controlled lattice modulation strategies, particularly via high-valence element doping which effectively stabilizes the crystal framework through strong bonding characteristics and charge compensation effects.

A University of Iowa-led research team has documented in humans that physical exercise sparks an increase in brain waves called ripples connecting areas in the brain linked to learning and memory. The researchers noted that a single exercise session spawns a spike in ripples and learning-memory connections. Results published in the journal Brain Communications.

Patients with Type 1 diabetes (T1D) require accurate and consistent monitoring of their blood glucose levels. Over the past decade, AI models have been explored to tackle this challenge; however, inter-patient variability and large data volumes remain key challenges. In a new study, researchers present BiT-MAML, a model-agnostic algorithm aimed at personalized blood glucose prediction of patients with T1D. This approach overcomes the limitations of existing models and enables precise predictions in real clinical settings.