Smartwatches commonly use heart rate variability to monitor stress. Columnist Helen Thomson explores what this metric actually tells us, and whether it could also predict and diagnose depression – and help improve your mental health more generally.
Get the latest international news and world events from around the world.
Frontiers: Introduction:
Alzheimer’s disease, a progressive neurodegenerative disorder, is marked by beta-amyloid plaque accumulation and cognitive decline. The limited efficacy and significant side effects of anti-amyloid monoclonal antibody therapies have prompted exploration into innovative treatments like focused ultrasound therapy. Focused ultrasound shows promise as a non-invasive technique for disrupting the blood–brain barrier, potentially enhancing drug delivery directly to the brain and improving the penetration of existing therapeutic agents.
Mars dust storms are sparking electricity and rewriting the planet’s chemistry
Mars may look like a quiet, dusty world, but it’s actually buzzing with hidden electrical activity. Powerful dust storms and swirling dust devils generate static electricity strong enough to spark faint glowing discharges across the planet, triggering chemical reactions that reshape its surface and atmosphere. Scientists have now shown that these tiny lightning-like events can create a surprising mix of chemicals—including chlorine compounds and carbonates—and even leave behind distinct isotopic “fingerprints.”
Mars is often portrayed as a dry, lifeless desert, but it is far more active than it appears. Its thin atmosphere and dusty terrain create an environment where constant motion generates electrical energy. Dust storms and spinning dust devils sweep across the surface, continually reshaping the landscape and driving processes that scientists are only beginning to fully understand.
Planetary scientist Alian Wang has been studying this phenomenon in depth. In a series of studies, including recent work published in Earth and Planetary Science Letters, she has examined how these electrically charged dust activities influence the chemistry of Mars, particularly through their impact on isotopes.
Solar reactor uses old battery acid to turn plastic waste into clean hydrogen
Researchers have developed a solar-powered reactor to break down hard-to-recycle forms of plastic waste—such as drink bottles, nylon textiles and polyurethane foams—using acid recovered from old car batteries, and converting it into clean hydrogen fuel and valuable industrial chemicals. The results are reported in the journal Joule.
The reactor, developed by researchers from the University of Cambridge, is powered by the energy from the sun, and could be a cheaper, more sustainable alternative to current chemical-based recycling methods. The team says their method could create a circular system where one waste stream solves another.
Global plastic production is more than 400 million tons per year, yet only 18% is recycled. The rest is burned, landfilled, or leaks into ecosystems. The researchers believe that their method, known as solar-powered acid photoreforming, could become part of the solution to the global mountain of plastic waste.
Aquila Booster turns a weak pulsar into a powerhouse of PeV particles
A point-like cosmic particle accelerator pumps out PeV gamma rays stronger than expected from a pulsar 50x weaker than Crab.
What makes this discovery remarkable is not just the energy, but the efficiency. This system appears to convert energy into high-speed particles far more effectively than current physics says it should.
In simple terms, astronomers may have found a cosmic particle accelerator that outperforms even their best theoretical designs.
To understand the breakthrough, it helps to know what scientists were looking at. A pulsar wind nebula forms when a dead star, called a pulsar, spins rapidly and blasts out a stream of charged particles at nearly the speed of light.
The fake disease that fooled the internet, and what it says about all of us
Until a few years ago, no one had heard of bixonimania. Then, in 2024, a group of scientists posted findings online announcing the condition, which they claimed affected the eyes after computer use. However, the scientists had made it up—not just the work, but the authors’ names, affiliations, locations and funding, which was the University of Fellowship of the Ring and the Galactic Triad.
Large language models like ChatGPT and Gemini treated it as real anyway, and in doing so, helped turn a fictional disease into a legitimate-sounding health concern.
Bixonimania is not an isolated case. Being deceived—whether you are a person or an AI model—is concerningly common, in science and beyond. Whether we’re talking about AI hallucinations, state-backed disinformation or just everyday lies, humans have a remarkable knack for naivety, owing to our biases and increasing need to outsource learning to others. These are problems we—individually and collectively—urgently need to better understand and overcome.
New deadly disease outbreak map flags ‘highly vulnerable’ regions around the world
New global modeling shows that about 9.3% of the world’s land area is highly vulnerable to the risk of dangerous disease outbreaks.
These hotspots are concentrated in Latin America and Oceania, where communities already face pressure from climate change and land development.
The research also identifies the countries most vulnerable to outbreaks – and the least equipped to detect and contain them.
Quantum-informed machine learning for predicting spatiotemporal chaos with practical quantum advantage
Ultimately, QIML proves that we don’t need a fully fault-tolerant quantum computer to see results. By using quantum processors to learn the complex “rules” of chaos, we can give classical computers the boost they need to make reliable, long-term predictions about the most turbulent environments in the natural world.
Modeling high-dimensional dynamical systems remains one of the most persistent challenges in computational science. Partial differential equations (PDEs) provide the mathematical backbone for describing a wide range of nonlinear, spatiotemporal processes across scientific and engineering domains (1–3). However, high-dimensional systems are notoriously sensitive to initial conditions and the floating-point numbers used to compute them (4–7), making it highly challenging to extract stable, predictive models from data. Modern machine learning (ML) techniques often struggle in this regime: While they may fit short-term trajectories, they fail to learn the invariant statistical properties that govern long-term system behavior. These challenges are compounded in high-dimensional settings, where data are highly nonlinear and contain complex multiscale spatiotemporal correlations.
ML has seen transformative success in domains such as large language models (8, 9), computer vision (10, 11), and weather forecasting (12–15), and it is increasingly being adopted in scientific disciplines under the umbrella of scientific ML (16). In fluid mechanics, in particular, ML has been used to model complex flow phenomena, including wall modeling (17, 18), subgrid-scale turbulence (19, 20), and direct flow field generation (21, 22). Physics-informed neural networks (23, 24) attempt to inject domain knowledge into the learning process, yet even these models struggle with the long-term stability and generalization issues that high-dimensional dynamical systems demand. To address this, generative models such as generative adversarial networks (25) and operator-learning architectures such as DeepONet (26) and Fourier neural operators (FNO) (27) have been proposed. While neural operators offer discretization invariance and strong representational power for PDE-based systems, they still suffer from error accumulation and prediction divergence over long horizons, particularly in turbulent and other chaotic regimes (28, 29). Recent work, such as DySLIM (30), enhances stability by leveraging invariant statistical measures. However, these methods depend on estimating such measures from trajectory samples, which can be computationally intensive and inaccurate in all forms of chaotic systems, especially in high-dimensional cases. These limitations have prompted exploration into alternative computational paradigms. Quantum machine learning (QML) has emerged as a possible candidate due to its ability to represent and manipulate high-dimensional probability distributions in Hilbert space (31). Quantum circuits can exploit entanglement and interference to express rich, nonlocal statistical dependencies using fewer parameters than their promising counterparts, which makes them well suited for capturing invariant measures in high-dimensional dynamical systems, where long-range correlations and multimodal distributions frequently arise (32). QML and quantum-inspired ML have already demonstrated potential in fields such as quantum chemistry (33, 34), combinatorial optimization (35, 36), and generative modeling (37, 38). However, the field is constrained on two fronts: Fully quantum approaches are limited by noisy intermediate-scale quantum (NISQ) hardware noise and scalability (39), while quantum-inspired algorithms, being classical simulations, cannot natively leverage crucial quantum effects such as entanglement to efficiently represent the complex, nonlocal correlations found in such systems. These challenges limit the standalone utility of QML in scientific applications today. Instead, hybrid quantum-classical models provide a promising compromise, where quantum submodules work together with classical learning pipelines to improve expressivity, data efficiency, and physical fidelity. In quantum chemistry, this hybrid paradigm has proven feasible, notably through quantum mechanical/molecular mechanical coupling (40, 41), where classical force fields are augmented with quantum corrections. Within such frameworks, techniques such as quantum-selected configuration interaction (42) have been used to enhance accuracy while keeping the quantum resource requirements tractable. In the broader landscape of quantum computational fluid dynamics, progress has been made toward developing full quantum solvers for nonlinear PDEs. Recent works by Liu et al. (43) and Sanavio et al. (44, 45) have successfully applied Carleman linearization to the lattice Boltzmann equation, offering a promising pathway for simulating fluid flows at moderate Reynolds numbers. These approaches, typically using algorithms such as Harrow-Hassidim-Lloyd (HHL) (46), promise exponential speedups but generally necessitate deep circuits and fault-tolerant hardware.
Quantum-enhanced machine learning (QEML) combines the representational richness of quantum models with the scalability of classical learning. By leveraging uniquely quantum properties such as superposition and entanglement, QEML can explore richer feature spaces and capture complex correlations that are challenging for purely classical models. Recent successes in quantum-enhanced drug discovery (37), where hybrid quantum-classical generative models have produced experimentally validated candidates rivaling state-of-the-art classical methods, demonstrate the practical potential of QEML even before full quantum advantage is achieved. Despite these strengths, practical barriers remain. QEML pipelines require repeated quantum-classical communication during training and rely on costly quantum data-embedding and measurement steps, which slow computation and limit accessibility across research institutions.
