Researchers at the Mark and Mary Stevens Neuroimaging and Informatics Institute (Stevens INI) at the Keck School of Medicine of USC have developed a groundbreaking brain imaging technique that reveals how tiny blood vessels in the brain pulse with each heartbeat—changes that may hold clues to aging and diseases such as Alzheimer’s. The study, published in Nature Cardiovascular Research, introduces the first noninvasive method for measuring “microvascular volumetric pulsatility”—the rhythmic expansion and contraction of the brain’s smallest vessels—in living humans. Using ultra-high-field 7T magnetic resonance imaging (MRI), the team showed that these microvessel pulses increase with age, especially in the brain’s deep white matter, a region critical for communication between brain networks that is particularly susceptible as people age to reduced blood supply of distal arteries, the blood vessels that carry blood away from the heart and into the farthest parts of the body. Increasing microvessel pulses can disrupt systems in the brain, possibly speeding up memory loss and Alzheimer’s disease. The USC team’s innovation combines two advanced MRI approaches—vascular space occupancy (VASO) and arterial spin labeling (ASL)—to track subtle volume changes in microvessels over the cardiac cycle. The researchers confirmed that older adults show heightened microvascular pulsations in deep white matter compared to younger adults, and that hypertension further amplifies these changes.

A team led by Prof. LIU Fanglin from the Hefei Institutes of Physical Science of the Chinese Academy of Sciences has developed a novel method based on generative models for ecological soundscape analysis.

The article examines how machine learning is revolutionizing igneous petrology and volcanology by automating tasks, enhancing models, and accelerating discoveries. At the same time, the authors warn of key challenges, including the need to understand what models actually learn and to ensure transparency, reproducibility, and interpretability. These concerns are especially critical for volcanic hazard assessment and crisis management. The study also addresses ethical risks and reviews evolving policies in the EU, US, and China.

In a paper published on aBIOTECH, the authors developed a lightweight open-source deep learning algorithm automates wheat Fusarium Head Blight diseased spikelet rate measurement from phone-captured images. With 93.8% AP and 7.2M parameters, this algorithm is fit for deployment on mobile devices to facilitate resistance breeding.

We show that regionally endothermic shortfin mako sharks have enhanced thermoregulatory ability with an order higher warming rate than cooling rate during repeated dives, likely optimizing foraging efficiency. This ability, only comparable to bigeye tuna and swordfish in teleosts, represents the key adaptive significance of regional endothermy in fishes.