Frost is one of the most punishing adversaries of tea cultivation. A single cold snap can scorch tender leaves, derail an entire season’s harvest, and degrade the delicate chemistry that gives tea its value. For growers and breeders, however, judging how badly a field has been damaged has long been a stubbornly subjective exercise: visual ratings lump continuous injury into crude categories, and laboratory assays destroy the very tissue they measure. A new study published in Plant Methods by Zhengtong He of Shihezi University, Zhiwei Chen of the Tea Research Institute of Shandong Academy of Agricultural Sciences, and colleagues offers a way out, pairing hyperspectral imaging with a purpose-built deep learning model to read frost damage directly from the way leaves reflect light.

The core problem the team confronted is that frost injury is not a binary event but a continuum. Leaves exposed to sub-zero temperatures suffer a cascade of overlapping stresses: photosynthetic pigments degrade, the machinery of photosynthesis falters, and cell membranes and walls break down. Any single measurement captures only a sliver of that cascade. To capture the whole picture, the researchers collected hyperspectral data together with eight key physicochemical indicators from tea plants growing across natural frost damage gradients in the field, rather than in the artificial conditions of a growth chamber. From these multidimensional physiological responses they constructed a composite measure, the tea cold stress response index, or TCSRI, which expresses frost damage severity as a continuous value instead of a coarse visual class.

Building TCSRI is a meaningful advance in itself. Traditional frost assessment often relies on ordinal scales, where an observer assigns a score of mild, moderate, or severe. Such scales throw away information and make it difficult to track recovery or compare cultivars with fine precision. By integrating several physiological dimensions into one index, the team created a target variable that reflects the plant’s actual internal state. The challenge then became predicting that index, and the individual chemical indicators behind it, from spectral data alone, without touching or destroying a single leaf.

Hyperspectral sensing is well suited to this task in principle. Across hundreds of narrow wavelength bands, plant leaves encode a wealth of information: chlorophyll absorption in the visible range, water and structural features in the near infrared. But the data are high-dimensional, with far more spectral channels than samples, a regime that invites overfitting, especially in field studies where collecting ground-truth measurements is laborious. Conventional chemometric approaches, from partial least squares regression to standard machine learning pipelines, often struggle to separate genuine physiological signals from noise, sensor drift, and the confounding effects of leaf surface properties and illumination variation.

The researchers’ answer is a model they call STAMPformer, short for spectral-prior-guided multi-view attention pooling. Its architecture reflects a deliberate design philosophy: rather than letting a neural network rediscover everything from scratch, the model is steered by what spectroscopists already know about how plants respond to stress. At its front end, a parameter-free attention convolution extraction module jointly models both the raw reflectance spectra and their first-order derivatives. Derivative spectra are a classic spectral prior: they accentuate subtle local features such as absorption peaks and shoulders that are easily drowned out in raw reflectance, while suppressing broad baseline drift and noise. By processing both views together, the module sharpens the local spectral signatures that carry physiological meaning.

On top of this extraction stage sits the second innovation, a multi-view attention pooling module. Deep networks processing spectral sequences produce long streams of features, but not all wavelengths matter equally for a given prediction. The pooling module adaptively aggregates sequence features drawn from different feature subspaces, learning to amplify the responses that are most diagnostic of cold stress and to down-weight the rest. This attention mechanism acts like a learned spotlight, and because it operates across multiple subspaces, it can capture complementary patterns that a single attention head might miss. The result is a compact representation tailored to the prediction task at hand, whether that is estimating a specific chemical indicator or the composite TCSRI.

The performance numbers are striking. Across all nine prediction tasks, spanning the eight physicochemical indicators and the TCSRI itself, STAMPformer achieved coefficients of determination above 0.87 and ratios of performance to deviation above 2.80, thresholds generally regarded as indicating reliable quantitative prediction. In other words, from a leaf’s reflectance fingerprint alone, the model could reconstruct the plant’s internal frost-stress state with an accuracy that approaches laboratory measurement, while remaining entirely non-destructive. The team did not stop at headline metrics. Ablation experiments systematically removed each proposed module to confirm that both the attention convolution extraction and the multi-view pooling contributed measurably to the gains, and paired statistical tests across multiple random seeds ensured the improvements were not artifacts of a lucky initialization.

Interpretability was a further priority, and here the study delivers something rarer than raw accuracy. Applying SHAP analysis, a technique that attributes each prediction to the contributions of individual input features, the researchers identified the wavelengths the model relied on most heavily. These key spectral regions correspond to physiologically meaningful processes: pigment degradation, photosynthetic impairment, and the disruption of cellular structure. That alignment between machine attention and plant biology matters on two fronts. It gives spectroscopists confidence that the model has learned genuine stress signatures rather than spurious correlations, and it hands plant physiologists a data-driven map of which parts of the spectrum track which facets of cold injury, potentially guiding future sensor design and breeding assays.

Crucially, the model also proved itself on data it had never seen. Independent external validation, using samples not involved in training, demonstrated that STAMPformer’s predictive ability generalizes beyond its original dataset, addressing one of the most common failure modes of machine learning in agricultural sensing. For a field where models are often tuned and tested on the same plots, this external check is an important credibility marker, suggesting the framework could be deployed across different farms, seasons, and frost events rather than remaining a laboratory curiosity.

The practical implications reach well beyond tea. An accurate, non-destructive, continuous measure of cold damage opens the door to rapid post-frost field surveys, letting growers quantify losses within hours instead of waiting for symptoms to develop or sending samples to a lab. For breeders screening for cold-tolerant germplasm, a spectral readout of TCSRI could accelerate the evaluation of hundreds of candidate lines without destructive sampling. And in the emerging vision of the smart tea garden, where drones and handheld spectrometers roam the rows, a model like STAMPformer could feed real-time frost maps into irrigation, shading, and recovery-management decisions. The authors frame their work as supporting frost monitoring, cold-tolerant germplasm evaluation, and intelligent garden management, and the architecture’s reliance on general spectral priors rather than tea-specific tricks suggests the approach could be adapted to other crops facing abiotic stress. As climate variability makes late-season frosts increasingly erratic, tools that turn light reflected from a leaf into a precise physiological diagnosis may become as essential to the tea garden as the thermometer.

Subject of Research: Hyperspectral deep learning prediction of continuous frost damage severity in tea plants

Article Title: STAMPformer: spectral-prior-guided multi-view attention learning for hyperspectral prediction of continuous frost damage severity in field-grown tea plants

Article References: He, Z., Chen, Z., Ding, Z., Wang, Z., Zhang, F., Zhao, Y., & Dong, C. (2026). STAMPformer: spectral-prior-guided multi-view attention learning for hyperspectral prediction of continuous frost damage severity in field-grown tea plants. Plant Methods. https://doi.org/10.1186/s13007-026-01597-4

Image Credits: AI Generated

DOI: 10.1186/s13007-026-01597-4

Keywords: tea plant, frost damage, hyperspectral imaging, deep learning, attention mechanism, cold stress index, non-destructive prediction, SHAP analysis, precision agriculture, spectral priors, plant physiology, smart farming

It’s easy to see why people might latch onto the idea of a reptilian brain as an explainer, or an explainaway, for complex topics.

Source

A German scientist, he developed a process for turning corpses into posable figures. His shows drew millions of visitors, and many detractors.

The mass extinction event known as the Great Dying, which occurred 252 million years ago, resulted in the loss of 96% of marine species, and for many years, scientists could not fully explain why certain animals disappeared while others survived.

That longstanding mystery may be closer to a resolution, thanks to a recent study in the Proceedings of the National Academy of Sciences, which suggests that differences in physiology, metabolism, and oxygen use helped determine which marine animals survived the catastrophe.

Stanford University researchers found that physiological differences largely determined which marine animals survived the end-Permian mass extinction after warming oceans lost much of their oxygen. Animals whose bodies could not meet rising oxygen demands, including brachiopods and crinoids that had long ruled the seafloor, had a hard time surviving as ocean temperatures rose and oxygen in the water dropped.

By contrast, mollusks and fish possessed body plans better suited to supplying oxygen as temperatures climbed, allowing them to survive in greater numbers.

“With this study, we essentially wanted to solve the mystery of why, when you go to the beach, you collect the shells of clams and snails rather than those of brachiopods,” said lead author Jose Andres Marquez, a former PhD student in the lab of Stanford’s Erik Anders Sperling.

Marquez and his team developed a model based on species’ metabolic oxygen balance and tested it using direct measurements of oxygen use in living relatives of both ancient groups. This included brachiopods collected from the San Juan Islands in Washington. Their results revealed that the risk of extinction closely aligned with how much oxygen a species required and how effectively it could transport oxygen within its body.

A Body Built for Survival

Modern sea animals, such as bivalves, gastropods, and fish, have become more active and muscular over time. Their hunting habits demand more oxygen, even under normal conditions. Although this could seem like a disadvantage during harsh times, it actually worked in their favor. When ocean temperatures rose and oxygen levels dropped during the Permian-Triassic period, these animals already relied on features such as gills, strong blood flow, and larger muscles to get more oxygen when needed.

On the other hand, Paleozoic animals like brachiopods and crinoids were slow-moving filter feeders that thrived in low-oxygen environments when things were stable. However, as temperatures rose, they required more oxygen, and their bodies could not keep up with the demand in the now-oxygen-poor waters.

“Our findings show that, across different organism groups, extinctions happened at much higher rates for those more vulnerable to increases in water temperature and decreases in oxygen availability,” Marquez said.

The findings support decades of evidence from the fossil record. Brachiopods, which once accounted for thousands of species, now number about 400. In contrast, bivalves have diversified and now account for between 10,000 and 15,000 species.

Settling the Debate

Sperling previously co-authored a 2018 study that identified oxygen loss and warming as the main causes of the Great Dying. However, that earlier research used only physiological data from modern species, so the results were influenced by data from commercially important fish and crustaceans rather than from the ancient groups that have largely disappeared. The new study addresses this limitation by directly measuring the physiology of living representatives of ancient animal groups, including brachiopods.

“This study is really the final nail in the coffin for what caused the Permian-Triassic mass extinction,” Sperling said.

A Familiar Climate

The environmental changes that led to the Great Dying began in oceans that were relatively cool and well-oxygenated, similar to present-day conditions. A massive release of volcanic carbon dioxide altered the atmosphere, driving global temperatures up by 8 to 12 degrees Celsius over several thousand years. By comparison, current climate projections estimate a rise of 1.5 to 4 degrees Celsius above preindustrial levels by 2100, but over a much shorter timescale. “The bad news is, we are on track for Permian-Triassic levels of warming in worst-case scenario projections,” Sperling said.

By comparing more groups of marine animals, the researchers hope to determine whether the physiological patterns that shaped Earth’s greatest mass extinction could help predict which species are most vulnerable as today’s oceans continue to warm. “We’re still at the point where we can change things and do something about it,” Sperling emphasized.

Austin Burgess is a writer and researcher with a background in sales, marketing, and data analytics. He holds an MBA, a Bachelor of Science in Business Administration, and a data analytics certification. His work focuses on breaking scientific developments, with an emphasis on emerging biology, cognitive neuroscience, and archaeological discoveries.

In a groundbreaking new study published in Pediatric Research, scientists have unveiled a novel biomarker within mothers’ milk that significantly advances our understanding of secretory activation in mothers of preterm infants. This discovery holds the potential to transform neonatal care by enabling earlier and more precise monitoring of lactation readiness, specifically through assessing sodium content without relying on the traditional sodium-to-potassium (Na:K) ratio. The implications for clinical practice and infant health are profound, especially for premature babies who depend on timely and adequate milk supply.

Secretory activation, the critical transition of the mammary gland to copious milk secretion, has long been a focal point for neonatal healthcare providers. Early and accurate detection of this physiological milestone can dramatically influence infant outcomes by ensuring that premature neonates receive the optimal nutrition needed for growth and immune support. Traditionally, the Na:K ratio in breast milk has served as the primary laboratory marker to signal secretory activation. However, this method presents challenges—samples must be analyzed carefully, and fluctuating electrolyte concentrations often complicate the interpretation, particularly in the dynamic physiology post-delivery.

The study spearheaded by Pei, Meier, Bendixen, and colleagues proposes a compelling alternative: monitoring the sodium (Na) concentration in breast milk independently, without referencing potassium (K) levels. By focusing solely on sodium levels, researchers discovered a more straightforward, faster, and potentially more sensitive biomarker to detect the onset of secretory activation. This shift could overhaul clinical protocols and streamline the path to identifying lactation progress in mothers of preterm infants.

Preterm birth, defined as birth before 37 weeks gestation, is a critical public health challenge affecting millions globally. Infants born prematurely are at elevated risk for complications stemming from immature organ systems, including underdeveloped gastrointestinal tracts that rely heavily on human milk to mitigate risks like necrotizing enterocolitis and infections. Timely secretory activation ensures that mothers can provide milk with adequate volume and composition tailored to meet these vulnerabilities, making the detection of this lactation stage an essential focus.

One of the key findings from the research reveals that sodium concentration alone, measured in mothers’ milk, reliably corresponds with the timing of secretory activation, offering enhanced accuracy over the Na:K ratio. The researchers harnessed cutting-edge biochemical assays and longitudinal sampling of milk from mothers of preterm infants, characterizing electrolyte fluctuations in detail. They demonstrated that sodium levels begin to decrease sharply after secretory activation due to the closure of tight junctions in mammary epithelium, making sodium a robust surrogate marker when potassium is disregarded.

This novel analytical approach holds significant clinical advantages. By simplifying the biomarker assessment to a single electrolyte—sodium—healthcare providers could deploy point-of-care testing devices that quickly inform about lactation status. This immediate feedback is not merely academic; it could empower bedside decision-making, maternal counseling, and targeted interventions such as lactation support or nutritional supplementation tailored to the mother-infant dyad’s needs.

Moreover, the researchers emphasized the particular relevance of this method for mothers of preterm infants, who face complex physiological challenges in initiating and maintaining milk production. The stress of prematurity, along with common maternal health complications, can delay secretory activation, placing these vulnerable infants at nutritional risk. Early detection of such delays via a straightforward sodium measurement could prompt timely remedial strategies, ultimately improving growth trajectories and developmental outcomes.

The molecular and cellular underpinnings lend biological credibility to these findings. Secretory activation involves hormonal cascades, primarily triggered by the precipitous drop in circulating progesterone after placental delivery, alongside rising prolactin levels. This hormonal milieu encourages the closure of mammary tight junctions, reducing sodium passage into milk. Hence, declining sodium concentrations mirror the biochemical shifts indicating functional milk secretion—transforming sodium into a powerful diagnostic proxy.

Beyond the immediate neonatal implications, this study’s methodology aligns with larger trends in personalized medicine and point-of-care diagnostics. By distilling complex physiological events into accessible, quantifiable milk component metrics, the research epitomizes the merging of biochemistry with clinical practicability. As lactation science evolves, such innovations hold promise to optimize maternal-infant health with greater precision and speed.

Nevertheless, the study authors acknowledge potential limitations and areas for further inquiry. The population was focused on mothers of preterm infants, and whether the sodium-only marker applies equally to term deliveries or diverse populations warrants exploration. Additionally, advancing from laboratory assays to user-friendly devices suitable for widespread clinical or even home use will require multidisciplinary collaboration in engineering, biochemistry, and healthcare delivery.

Intriguingly, the research opens doors to investigating how other milk components—such as immunoglobulins, hormones, or other electrolytes—could synergize with sodium assessment to map lactation physiology more comprehensively. Future studies might expand this framework to encompass not just secretory activation timing but also milk quality, volume forecasting, and infant growth correlation, painting a more holistic picture of lactation dynamics.

The implications extend beyond immediate clinical management. For neonatologists and lactation consultants, a reliable and rapid biomarker offers a potent tool for enhancing support programs, tailoring interventions, and potentially improving breastfeeding success rates and duration, especially in the fragile realm of neonatal intensive care. With breastfeeding recognized globally as a cornerstone of infant health, innovations like these resonate far across public health landscapes.

In sum, Pei and colleagues’ work represents a pivotal advancement in lactation science and neonatal nutrition. By elucidating the power of sodium measurements without the confounding Na:K ratio, they have charted a path toward more streamlined, accurate, and clinically actionable monitoring of secretory activation in mothers of preterm infants. This biomarker could become a vital asset in neonatal care, ensuring the earliest possible identification of lactation onset and facilitating optimal support for both mother and child during a crucial period.

As this research garners attention, it is poised to influence future guidelines and inspire technological innovation in lactation monitoring. The prospect of employing simple sodium assays in diverse care settings—from tertiary hospitals to community clinics—heralds a democratization of lactation science, empowering more mothers to achieve successful breastfeeding outcomes, regardless of prematurity challenges.

The cascade of benefits from this research is already rippling through the scientific and medical communities. Improved neonatal nutrition, reduced complications from feeding delays, and enhanced maternal confidence all emerge from a seemingly small shift in how we understand and measure breast milk composition. Such breakthroughs underscore the power of focused biochemical insight to generate transformative healthcare innovations.

Looking forward, collaborations between researchers, clinicians, and technologists will be essential in translating this discovery into effective clinical tools. As innovation speeds up, the integration of rapid sodium tests into neonatal intensive care workflows could soon become standard, ensuring that no preterm infant is left behind in the race toward optimal growth and development. This study not only advances science but also embodies a vital step toward better health outcomes worldwide.

---

Subject of Research: Secretory activation in mothers of preterm infants as detected through sodium concentration measurements in breast milk.

Article Title: Mothers’ milk Na without Na:K ratio detects secretory activation in mothers of preterm infants.

Article References:
Pei, Q., Meier, P.P., Bendixen, M.M. et al. Mothers’ milk Na without Na:K ratio detects secretory activation in mothers of preterm infants. Pediatr Res (2026). https://doi.org/10.1038/s41390-026-05206-x

Image Credits: AI Generated

DOI: 19 June 2026

Scientists discovered a hidden gut-brain network that can change cravings when the body needs more protein. The body has a remarkable ability to recognize when it is missing important nutrients and adjust behavior accordingly. New research shows that the gut plays a much larger role in this process than previously understood, actively detecting protein shortages [...]