Big Health Inc, along with paid academic investigators, reports higher remission rates and lower anxiety symptom scores with their smartphone-delivered digital cognitive behavioral therapy, DaylightRx, compared with an online psychoeducation, also created by Big Health Inc.

Generalized anxiety disorder involves excessive, persistent, and uncontrollable anxiety with lifetime prevalence reported as 6%, alongside reduced quality of life, impaired social and occupational functioning, and increased health care utilization.

Cognitive behavioral therapy and pharmacotherapy are considered first-line treatments. Despite strong tolerability, efficacy, and cost-effectiveness, access remains limited due to a shortage of trained therapists, time burdens, and stigma.

Digital delivery has been proposed to address access barriers by expanding reach. Smartphones offer real-time access and near-universal availability. Meta-analytic reviews of early smartphone interventions for anxiety suggest limited effectiveness, prompting interest in more structured digital applications that incorporate core elements of cognitive behavioral therapy.

DaylightRx delivers audio-guided exercises and interactive instruction in cognitive restructuring, applied relaxation, stimulus control, avoidance reduction, mindfulness, problem solving, and imaginal exposure. All content is accessed via smartphone, with tools designed for everyday use through regular engagement.

In the study, “Digital Cognitive Behavioral Treatment for Generalized Anxiety Disorder,” published in JAMA Network Open, researchers tested a smartphone-based therapy called DaylightRx against a psychoeducation control in adults diagnosed with generalized anxiety disorder.

Trial procedures were fully remote and coordinated by Boston University. The cohort included 351 participants with generalized anxiety disorder confirmed by a structured interview, and scores of 15 or higher on the 7-item Generalized Anxiety Disorder scale. Screened-out conditions covered moderate or greater suicide risk, bipolar disorder, psychosis, obsessive compulsive disorder, and recent substance use disorder.

Control participants received written online psychoeducation covering symptoms, prevalence, causes, consequences, and lifestyle advice, with self-paced access and encouragement to apply material.

Since the psychoeducation control was created by Big Health Inc, the sponsor of the study and the creator of DaylightRx (the intervention the control is meant to gauge the effectiveness of), any result would only apply to this specific scenario and should not be generalized.

Clinical Global Impressions–Improvement (CGI-I) scale and self-reported anxiety severity with the 7-item Generalized Anxiety Disorder (GAD) scale, assessed at 10 weeks and 24 weeks after randomization. “Remission” used a Clinical Global Impressions–Improvement score of 2 or lower. Primary outcomes used blinded independent evaluator ratings for “remission.”

Hitting pause on ‘remission’

Remission is in quotes for this article, though it was not in the study, as this term was redefined internally by the study design. Remission typically refers to a state of minimal or absent symptoms, sustained over time, and often confirmed via structured diagnostic reassessment.

“Remission” is used in two distinct ways across the trial without consistent distinction. The primary outcome defines remission using a CGI-I score ≤ 2, based on evaluator judgment of improvement. A secondary outcome defines remission as a GAD-7 score <10 at week six, a self-reported symptom cutoff. Neither definition involves clinical reassessment or functional recovery metrics.

This dual use creates terminological ambiguity and may inflate the clinical impression of symptom change, particularly where “remission” might imply a diagnostic resolution.

Operational meaning differs by context within the report, and makes the use of “remission” more of a result of a branding effort that borders on misleading lay readers into believing the treatment is far more effective than evidenced by the study.

Unpausing to see the results

“Remission” at 10 weeks occurred in 71.0% of DaylightRx digital cognitive behavioral therapy participants with video visit data, compared with 34.6% of Big Health psychoeducation participants. “Remission” at 24 weeks occurred in 77.7% of digital cognitive behavioral therapy participants with video visit data, compared with 52.0% of their psychoeducation participants.

The mean 7-item Generalized Anxiety Disorder scale score at 10 weeks was 7.88 for digital cognitive behavioral therapy and 11.68 for psychoeducation. The mean score at 24 weeks was 7.23 for digital cognitive behavioral therapy and 10.68 for psychoeducation.

Adverse events

Events potentially related to digital cognitive behavioral therapy use included panic attacks at 2.9%, depression symptoms at 2.9%, hormonal or seasonal mood symptoms at 1.7%, suicidal ideation at 1.7%, posttraumatic stress disorder symptoms at 0.6%, musculoskeletal pain at 3.4%, and headache at 1.7%.

A number of unintended unblinding events occurred, all in the digital cognitive behavioral therapy arm, with sensitivity analyses described as robust when excluding those participants. Numerical results of the sensitivity analyses, or how the authors defined “robust,” were also excluded.

Implications for care

Researchers concluded that smartphone-delivered digital cognitive behavioral therapy offered significant and sustained benefits to adults with generalized anxiety disorder, with advantages over self-directed online psychoeducation at 10 weeks and 24 weeks.

Validation of a digital cognitive behavioral therapy program intended for prescribing by mental health and primary care practitioners would require a trial against established criteria and efficacy would need comparisons to existing validated methods, none of which occurred in the current study, leaving any unbiased conclusions unclear.

Use of “remission” as a clinical-sounding endpoint creates the appearance of psychiatric resolution, which may not be warranted by a drop in score alone. Readers unfamiliar with measurement-based care may misinterpret this as diagnostic remission, rather than symptom score reduction.

As sea levels rise and soils grow saltier, even the iconic Sabal palmetto—the official state tree of Florida also known as the cabbage palm—may be in danger. But a new study in HortScience suggests a simple, widely available amendment could help young palms survive in salty conditions.

Graduate student researcher, Pedro Gonzalez of Conservation and Sustainable Horticulture Lab and the team of researchers: Amir Khoddamzadeh (Florida International University Earth and Environment Department Institute of Environment), Patrick Griffith (Montgomery Botanical Center in Coral Gables), and Madhugiri Nageswara-Rao (USDA ARS Subtropical Horticulture Research Station in Miami) found that treating young cabbage palm seedlings with small amounts of silicon greatly improves their tolerance to salt stress.

With its sponge-like ability to retain water, silicon —a powerful ally known for its resilience-boosting properties, offers a lifeline for vulnerable young palms by improving physiological processes. Already widely used—even by NASA to grow plants in space—it’s both commercially accessible and affordable, making it a perfect candidate for helping palms survive salinity.

“We wanted to work on something that ornamental plant nursery producers and landscapers can use,” Khoddamzadeh said.

The team grew 96 one-year-old seedlings in a greenhouse for 12 months, watering them with varying salt levels, adding silicon to the soil in different amounts and tracking everything from soil nutrients, plant growth and development to survival rates.

The study also established the ecological threshold for seedling survival—a critical finding for future restoration efforts.

The results show that salinity alone dramatically reduced growth, chlorophyll content, and leaf production, particularly at higher salt concentrations. At 50 ppt salinity (comparable to seawater), untreated seedlings died.

However, when silicon was added, seedlings at moderate salinity (10–30 ppt) maintained higher chlorophyll content, better leaf production and improved overall vigor. Even under more extreme salinity, silicon-treated seedlings partially preserved performance—though survival at the highest salinity remained low.

The implications are serious: cabbage palms are not only a cultural and ecological icon, but they support a multibillion‑dollar landscaping and nursery industry in Florida, generating thousands of jobs.

“Even in all its glory, this quintessential giant is not spared from the growing threats of sea‑level rise and salt exposure,” Khoddamzadeh said.

The researchers say that using silicon is both cost‑effective and simple—just a few grams mixed with water can help young palms survive.

Beyond nurseries, the findings may guide restoration efforts: communities might plant adult palms (more salt‑tolerant) in safer locations or choose native species that better resist salinity.

As South Florida braces for rising seas and increasing saltwater intrusion, this study offers a practical tool to help preserve a symbol of the state.

Most people have either been told that they don’t look well when they were sick, or thought that someone else looked ill at some point in their lives. People often use nonverbal facial cues, such as drooping eyelids and pale lips, to detect illness in others, potentially to prevent infection in themselves. A new study, published in Evolution and Human Behavior, finds that women are more sensitive to these subtle cues than men.

Rating the natural signs of illness

In past studies, participants have been asked to rate signs of illness in the faces of others, but some of these studies used manipulated photos or people who had artificially induced sicknesses in the photos. In the new study, the team wanted to see whether naturally sick individuals would be rated as sick-looking, or as having an expression of “lassitude,” by other individuals and whether the recognition differed by sex.

To do this, the team recruited 280 undergraduate students, of which 140 were male and 140 were female, to rate 24 photos. The photos consisted of 12 different faces in times of sickness and health.

The ratings were based on six illness-related dimensions, including: safety, healthiness, approachability, alertness, social interest and positivity, using 9-point Likert scales. These differing dimensions helped the researchers assess things like whether the participants felt as if they could approach the person in the photo and whether they looked like they were happy or tired, as well.

“Given that these dimensions are positively correlated with each other and have been previously used to assess sick face sensitivity, we created a latent lassitude perception variable, indexed by the six dimensions that each tapped unique but related constructs. We also predicted that sex would predict latent lassitude perception, with females showing more accuracy in their ability to discriminate between sick and healthy faces than males,” the study authors explain.

After analyzing the participants’ ratings they found that their hypothesis was correct—women, on average, were more sensitive to signs of illness in faces. The difference was small, but still statistically significant and consistent throughout the study.

Potential evolutionary roots of illness perception

There are two dominant hypotheses as to why women might be capable of detecting illness more accurately. The first is referred to as the “primary caretaker hypothesis,” which posits that because, throughout history, women were more often the ones taking care of infants and young children, they evolved to detect illness better. In theory, recognizing the nonverbal cues of illness would help women to detect illness in babies and young children faster. Ultimately, this ability increases the survival of their offspring.

Another hypothesis is the “contaminant avoidance hypothesis.” This hypothesis states that females experience higher levels of disgust compared to males.

The study authors write, “These differences are theorized to result from repeated periods of immune suppression across the reproductive lifespan, occurring both during pregnancy and in the luteal phase of the monthly cycle, in anticipation of pregnancy. Females, therefore, overall, may have had greater selective pressure for disease-avoidance than males.”

The researchers note that this study was limited to undergraduate students, which may not generalize to broader populations. Also, indicators of sickness, like voice and posture, were not included. The photos in the study were only of stationary, cropped faces. These additional indicators may influence sickness perception to a different degree.

Putting solar panels above agricultural crops may do more than produce food and clean energy on the same land: It can also significantly augment quality of life for farmworkers, according to new research to be presented at AGU’s 2025 Annual Meeting in New Orleans. Worker-reported benefits include shelter from the sun, cooler drinking water and reduced fatigue, while physical measurements indicate the panels can help farms avoid conditions conducive to dangerous heat stress.

“In a lot of [food] sustainability conversations, we’re thinking about resource use and not about farmworkers and their bodies,” said Talitha Neesham-McTiernan, a human-environment researcher at the University of Arizona who led the research. She will present her work on 15 December at AGU25, joining more than 20,000 scientists discussing the latest Earth and space science research.

A bundle of overlooked, but crucial, benefits

Hybrid solar-food fields, better known as “agrivoltaics” systems, typically involve solar panels mounted at or above head height, spaced among crops to allow sunlight to pass through the gaps between. In addition to making efficient use of land, these systems can benefit crops by reducing both sun damage and water lost to evaporation—and even by trapping some heat near the ground during colder months, Neesham-McTiernan said.

In her four years of fieldwork on farms like these, often during brutal Arizona summers, Neesham-McTiernan noticed a pattern: Researchers and farmworkers alike would strategically plan to work in the panels’ shade during the hottest hours.

“It just seemed to be something that people in these systems were doing, but nobody in the research area was talking about it,” she said. That struck her as odd, as farmworkers are 35 times more likely to die from heat-related illness than non-agricultural workers. With climate change pushing that figure higher, making any tool to reduce heat stress would be increasingly valuable.

To end that silence, Neesham-McTiernan and her co-authors asked seven full-time farmworkers at Jack’s Solar Garden, a small agrivoltaics farm near Longmont, Colorado, how their experiences differed from those on traditional farms.

The biggest reported perk, by far, was shade. One worker, Neesham-McTiernan said, confessed they found it hard to imagine ever going back to work on traditional full-sun farms—where, they added, their favorite crops had always been tomatoes, because of the shade the tall plants offered.

“By 9 a.m., in the summer, you’re just cooking,” Neesham-McTiernan said. “Being able to take that direct heat load off makes such a difference.”

Shade keeps drinking water cool too, the workers noted—a crucial benefit, given water’s role in mitigating heat stress. “They can pop their bottles under the panels and they stay cool all day,” Neesham-McTiernan said, “rather than it being, as one of the farmworkers described it, like drinking tea.”

Another worker said these benefits helped them feel less exhausted by day’s end, leaving more energy for social life and allowing a faster recovery for the next day’s work. Others said simply knowing shade was nearby reduced their mental stress.

To tell the full story of heat stress, gather stories and numbers alike

The researchers also recorded air temperature, wind speed, humidity and solar radiation to quantify heat stress metrics such as wet bulb globe temperature, which is commonly used to identify dangerous outdoor work conditions.

Compared to open-field farms, they found, agrivoltaics reduced wet bulb globe temperature by up to 5.5 degrees Celsius (10 degrees Fahrenheit)—the difference, Neesham-McTiernan estimates, between stop-work conditions and simply requiring a break every hour. “When that builds up over a day, over a season, over a lifetime of harvesting, that’s really significant.”

That’s not to say the measurements always matched farmworkers’ testimonies: for instance, they occasionally disagreed over which parts of the farm were hottest at which times of day. But fully understanding the experience of heat stress, Neesham-McTiernan said, requires both personal and measured evidence.

“Every farmworker said one benefit was being able to lean against the beams that hold up the panels, just to take the weight off a bit,” she noted. “If I just had my sensors in the field, I wouldn’t know that, but it clearly makes such a difference in their day-to-day comfort.”

Neesham-McTiernan said she’s working to expand the research into other regions to see whether the benefits apply in different environments. She also hopes to eventually collect more rigorous physiological and health data to quantify the impacts of agrivoltaics on workers’ bodies.

“[Agrivoltaics] isn’t a one-size-fits-all solution,” she said. “It can’t be used everywhere. But with the threat of heat, we need a catalog of ways we can protect farmworkers. Without them, we can’t feed ourselves. Protecting them and their bodies should be paramount to everyone.”

A report from analysts at Jefferies suggested that new screenings for metachromatic leukodystrophy and Duchenne muscular dystrophy could bump sales of the gene therapy Libmeldy by more than $100 million.

Health Secretary Robert F. Kennedy Jr.’s recent move to add two new conditions to standard newborn screening protocols could put money in Kyowa Kirin’s coffers, according to a report from Jefferies analysts published Tuesday.

Earlier this week, Kennedy added metachromatic leukodystrophy (MLD) and Duchenne muscular dystrophy to the list of diseases included on the federal government’s Recommended Uniform Screening Panel (RUSP), a list of diseases that the Department of Health and Human Services (HHS) suggests that newborns be tested for.

While states are not bound by the list, Jefferies anticipates that Kyowa, which sells a one-time gene therapy for MLD called Libmeldy through its subsidiary Orchard Therapeutics, could benefit. Adding MLD to the RUSP “is likely to accelerate the adoption of MLD screening and boost demand for the gene therapy treatment, Libmeldy,” the analysts wrote, noting that the treatment carries a price tag of $4.25 million.

Finding More Patients

The FDA approved Libmeldy in 2024 and Orchard began selling it in the U.S. this year. Jefferies estimated that four patients in the country will be treated this year, amounting to ¥2.1 billion ($13.3 million). With the change to RUSP, Jefferies suggested that at least 14 states, accounting for more than half of the country’s population, will screen for MLD, possibly bumping the total treated in the U.S. to 40 each year. That could add ¥18 billion ($114.3 million) in sales for Libmeldy.

Jefferies estimated that worldwide, Libmeldy’s sales will hit ¥6.9 billion ($43.8 million) this year.

Patients with MLD, a rare lysosomal storage disease that effects the nervous system, most often die between ages five and eight, and though Libmeldy doesn’t completely block symptoms, it does extend survival and slow progression when administered early, according to the Jefferies report.

Orchard commended the recommended inclusion of MLD on the list in a statement released Tuesday. CEO Bobby Gaspar cautioned, however, that while “the addition of MLD to the RUSP is a monumental step toward enabling newborn screening in the U.S., implementation now must happen at the state-level, a historically multi-year process inhibited by the lack of adequate federal and state funding.”

Debra Miller, the founder and CEO of research and advocacy group CureDuchenne, was pleased with Kennedy’s announcement as well. “We applaud HHS for this decision, which marks a breakthrough moment for our community,” she said in a statement to BioSpace. “Early diagnosis offers families a critical window to access care and emerging treatments when they may be most effective.”

Libmeldy is the only FDA-approved treatment for children with MLD on the market. It provides patients with a functional copy of the ARSA gene, which is mutated in the condition.

“This approval represents important progress in the advancement and availability of effective treatments, including gene therapies, for rare diseases,” then-director of the Office of Therapeutic Products Nicole Verdun said at the time of the therapy’s approval.

BioMarin Pharmaceutical has faced a rocky road, promising and then backing off revenue targets and cutting assets that have underperformed. But Amicus’ rare disease portfolio is already bringing in $600 million annually.

BioMarin Pharmaceutical is purchasing Amicus Therapeutics in an all-cash deal worth $4.8 billion, in a bid to pump up its rare disease portfolio after a series of cuts.

The deal prices shares of the New Jersey–based Amicus at $14.50, a 33% premium on the company’s last closing value, according to BioMarin’s announcement Friday morning. Shares of BioMarin are up 4.6% to $54.34 in pre-market trading Friday.

The acquisition brings two FDA-approved drugs to BioMarin’s rare disease offerings: Galafold, an oral treatment for Fabry disease, and Pombiliti/Opfolda, a two-part therapy for Pompe disease. Amicus also brings the U.S. rights to DMX-200, a Phase III small molecule treatment for focal segmental glomerulosclerosis, a rare and fatal kidney disease.

BioMarin hopes the Amicus buy will “accelerate revenue growth” for the company. Over the past year, Galafold and Pombiliti/Opfolda together generated $599 million in total sales for Amicus, according to the deal announcement.

The transaction will be a shot in the arm for BioMarin if those sales figure hold. In fall 2024, BioMarin promised investors $4 billion in annual revenue by 2027, without accounting for potential mergers and acquisitions.

By October of this year, the company had abandoned that hope, cutting revenue guidance for its achondroplasia therapy Voxzogo. That treatment was one of the company’s top sellers in 2024, with $735.1 million in revenue. It fell short only to Vimizim, BioMarin’s mucopolysaccharidosis treatment, which brought in $739.8 million for the year, according to the company’s annual report.

When announcing the $4 billion goal miss, BioMarin revealed the divestiture of hemophilia A treatment Roctavian, citing stagnant sales. Roctavian was supposed to be a major leg of the company’s revenue target. After a restructuring effort announced in late 2024, BioMarin reorganized itself into three units, one of which focused solely on Roctavian.

Cutting Roctavian was the second portfolio prune for BioMarin in recent months. In August, the company cut a preclinical phenylketonuria asset.

Amicus is the second major acquisition for BioMarin this year, after the purchase of enzyme replacement company Inozyme for $270 million in May. The revenue prospects from that purchase are a bit further out. Inozyme’s lead molecule, INZ-701 for treating the rare, progressive disease ectonucleotide pyrophosphatase/phosphodiesterase 1 (ENPP1) deficiency, just wrapped a Phase I/II trial late last year and is headed into late-stage development.

A compound that enables the delivery of insulin through the skin has been demonstrated in mice and minipigs. The findings, reported in a paper published in Nature, suggest a potential alternative to injection for diabetes management and may support broader applications in other therapeutics.

Drug delivery through skin is widely used for small molecules owing to its convenience and patient compliance. However, the structure of the skin presents a barrier to larger molecules such as proteins and peptides. Existing methods to enhance skin permeability, including microneedles, ultrasound, and chemical agents, are often invasive and compromise skin integrity, which limits their clinical utility.

Rongjun Chen, Youqing Shen, Jiajia Xiang, Ruhong Zhou and colleagues report a fast skin-permeable polymer called poly2-(N-oxide-N,N-dimethylamino)ethyl methacrylate, which can penetrate through different layers of skin through interactions with the skin’s changing pH gradient.

When combined with insulin, OP facilitates its transport through the skin into systemic circulation and its accumulation in key glucose-regulating tissues, including the liver and skeletal muscles. In diabetic mice and minipigs, application of OP–insulin to the skin lowered blood glucose levels to the normal range within 1–2 hours, which is comparable to the effect of injected insulin, and maintained the normal level for up to 12 hours. No adverse effects were observed in skin cells, blood cells or in the functions of organs including the liver and kidney.

The study demonstrates that OP could achieve skin permeation without disrupting skin structures, and that insulin combined with OP maintains its biological activity. Although further investigation is needed to assess long-term safety, dose control and potential clinical application, the strategy may offer a novel and versatile platform for noninvasive delivery of other biomacromolecules.

Trials of a new cholesterol-lowering pill have shown promising results for people with heterozygous familial hypercholesterolemia (HeFH), a genetic disorder that leads to high levels of LDL cholesterol.

HeFH is a common condition affecting about 1 in 250 people, caused by a mutation in a gene that impairs the body’s ability to remove low-density lipoprotein (LDL) cholesterol from the bloodstream. This inherited condition increases the risk for premature atherosclerotic cardiovascular disease (ASCVD)—a buildup of fatty deposits in arteries, leading to narrowed vessels that can restrict blood flow to vital organs.

The drug, called Enlicitide and developed by Merck, is a new type of PCSK9 inhibitor. It works by binding to PCSK9, a blood protein that typically degrades the liver receptors that clear LDL cholesterol. By blocking PCSK9, Enlicitide protects these receptors and boosts the liver’s ability to clear LDL cholesterol from the bloodstream, lowering the risk of heart disease.

The findings are published in the journal JAMA.

Year-long trial

The trial was a phase 3, 52-week, randomized trial that included 303 adults from 17 countries with HeFH who were already taking statins or other lipid-lowering therapies. Participants were randomly sorted into two groups. One group received the 20 mg Enlicitide pill once a day while the other group received an inactive pill (placebo). Neither the patients nor the doctors knew who received which.

After 24 weeks, LDL cholesterol levels dropped by an average of 58.2% in patients taking Enlicitide, while those on the placebo saw almost no change. At the 52-week stage at the end of the trial, Enlicitide patients achieved an average drop of 55.3% in their LDL cholesterol, while the placebo group saw their levels rise by 8.7%.

This potential new therapeutic also lowered levels of other cholesterol particles that contribute to ASCVD risk. Apolipoprotein B levels were reduced by 48.2% and Lipoprotein (a) levels decreased by 24.7%.

Drug safety

The drug was also well tolerated with few side effects. The proportion of participants reporting at least one adverse effect was similar between the groups: 77.7% for Enlicitide and 76.2% for the placebo. The proportion of participants who stopped taking the medication due to an adverse effect was also similar, with 2% for Enlicitide and 3% for the placebo.

“In adults with heterozygous familial hypercholesterolemia, Enlicitide is an effective and well-tolerated treatment for lowering the level of low-density lipoprotein cholesterol,” wrote the researchers in their study.

Ongoing trials of Enlicitide are gathering data about whether this powerful cholesterol reduction translates into fewer heart attacks and strokes. The scientists also want to test the pill in a wider population of high-risk patients beyond those with HeFH.

A new system could overhaul maps that misclassify hundreds of thousands of smallholder coffee and cacao farmers as working in forests. Without better maps, deforestation regulations could ripple through markets from remote farms to a caffe mocha near you.

Sample Earth, launched by the Alliance of Bioversity International and CIAT and available on Harvard Dataverse, helps mapmakers build accurate, inclusive maps to prevent smallholder farmers from being wrongly classified as producing major commodities in forested areas. Misclassification risks excluding compliant producers from markets enforcing deforestation-free rules, particularly the European Union’s new regulation (EUDR).

The initiative is the result of a collaboration between Alliance researchers, tech companies (including Google), and the World Cocoa Foundation. Researchers call on private-sector mapmakers to adopt their model to harden their supply chains against disruption.

Producers of coffee and cacao, and the companies that buy their products, could soon lose access to the world’s second-largest economy. The European Union, at the end of next year, will phase in the long-delayed EUDR legislation that requires many agricultural commodities to be certified deforestation-free. Unfortunately, hundreds of thousands of producers will face considerable hurdles, and not because they produce on land that hasn’t been deforested since 2020 (the EU’s cutoff date): It’s due to maps that wrongly classify their farmland as forest.

For example, the EU’s main reference map, published in 2025, misclassifies more than half the coffee production zones in Colombia, China, Guatemala and Mexico as forest, according to research by the Alliance of Bioversity International and CIAT. Similar reference maps have the same shortcomings. This is because these maps are “trained” on land-cover datasets that largely exclude remote areas cultivated by smallholders.

Improving these maps is urgent. To spark the creation of better maps, the Alliance recently launched Sample Earth, a trusted and inclusive global benchmark and reference dataset that accurately represents remote smallholder farms. The initial data tranche includes approximately 100,000 open-access, time-stamped geolocation points in Ghana and Vietnam. The countries are the second-largest producers of cacao and coffee, respectively.

“Maps are needed for due diligence, and buyers will likely steer clear of areas misclassified as ‘high risk’ for deforestation,” said Louis Reymondin, a data scientist at the Alliance. “With Sample Earth, we invite governments, companies, NGOs and research institutions to invest in expanding this inclusive, high-quality land-cover reference to preserve livelihoods and incentivize environmental protection.”

Smallholders produce an estimated 60% of the world’s coffee and 90% of its cacao. If maps used for compliance are inaccurate, buyers may decline purchases from entire regions rather than risk penalties for non-compliance, effectively shutting smallholders out of major markets.

“Most maps are not accurate at local scales because the data is biased toward regions with a lot of training data,” said Thibaud Vantalon, a scientist at the Alliance’s Digital Inclusion research area. “Remote regions are very poorly mapped. Sample Earth means to fill this gap in training data for smallholders.”

Making map-making better

Sample Earth is designed to improve map accuracy and to streamline the map-making workflow. Data scientists, the people who make maps with satellite imagery, spend an estimated 80% of their time collecting, cleaning and organizing training data. Sample Earth provides reference samples to reduce that burden and speed up the creation of accurate land-cover maps for compliance.

“High-quality data and data-based action are the foundation for compliance with deforestation-free rules and net-zero carbon emission targets,” said Michael Matarasso, the Impact Director and Head of North America at the World Cocoa Foundation (WCF), a partner in Sample Earth.

“However, highly accurate public data is rare… This poses a significant risk to all stakeholders involved. A standard to deliver highly accurate and transparent data in partnership with governments and farmers is of critical importance more than ever.”

Sample Earth aims to set a new transparency and quality benchmark for map-based compliance tools. Currently, no universal standard exists for third-party accuracy assessments of maps used in deforestation due diligence. Sample Earth plans to include a built-in improvement mechanism that allows mapmakers to access confidential land-use reference data to validate and refine their maps without exposing individual farmers’ locations.

“Global forest maps have advanced, but without open, standardized reference data, progress in disambiguating forest land use from other land use like cacao and coffee agroforestry remains limited” said Rémi d’Annunzio, Forestry Officer at FAO and product manager of Whisp. “Today, initiatives like the Forest Data Partnership and DIASCA are putting efforts such as Sample Earth high on the global agenda as we work to define and standardize guidelines for open reference data collection.”

Sample Earth builds on nearly two decades of Alliance research using satellite imagery to monitor land-cover changes across the Global South. The team plans to expand the dataset within Vietnam and Ghana and add other countries with high rates of misclassified smallholder farms, including Colombia and Honduras, along with coffee- and cacao-producing nations across Africa and Asia.

Seeking modern cartographers

Sample Earth’s roster of collaborators includes the United Nations’ Food and Agriculture Organization, Germany’s international development agency (GIZ), Google, Satelligence and WCF. The Alliance is actively seeking more collaborators and investors.

“For EUDR to succeed, we need to lower the burden of monitoring and reporting, and we need to ensure that longstanding smallholder farms can be reliably reported as non-deforested areas,” said Dan Morris, a researcher at Google AI for Nature and Society. “AI combined with satellite imagery is a powerful tool that can help address these challenges, but AI systems are only as good as their training and validation data.”

Inaction could disrupt supply chains and consumer markets, and not just in the EU; other jurisdictions are following suit in building similar legislation that will apply to most agricultural commodities. Supply constraints are feasible if maps do not quickly improve, which could push up prices. It’s bad news across supply chains, from vulnerable smallholders who already face myriad challenges to food-inflation-weary consumers worldwide.

Sample Earth’s proposition is straightforward: better, inclusive training datasets will yield more accurate maps, protect compliant farmers from unwarranted exclusion, and give buyers and governments transparent tools to verify deforestation-free claims. By filling the data gaps that leave smallholder landscapes underrepresented, Sample Earth aims to make compliance affordable and fair, while supporting conservation and sustainable livelihoods in the tropics.

Provided by The Alliance of Bioversity International and the International Center for Tropical Agriculture

Agricultural producers and manufacturers often need information about crop attributes, from nutrient content to chemical composition, to make management decisions. In recent years, multispectral imaging has emerged as a useful tool for product analysis, but the required equipment is expensive. Standard RGB cameras are much more affordable, but their images show only visible attributes.

However, if RGB images can be “translated” to multispectral images, pictures taken with a smartphone or any regular camera can yield sophisticated information. This process requires complex computer modeling and machine learning, but once the techniques are developed, they can be applied to simple devices anyone can use.

In two new papers published in Computers and Electronics in Agriculture, researchers at the University of Illinois Urbana-Champaign explore the reconstruction of multispectral and hyperspectral images from RGB for chemical analysis of sweet potatoes and maize.

“An RGB camera captures only the visible range in three bands, red, green, and blue. The pictures cannot provide any chemical information, which you often need for crop analysis. We reconstructed images from these three bands to include information from the near-infrared range, which you can use to determine chemical composition,” said Mohammed Kamruzzaman, assistant professor in the Department of Agricultural and Biological Engineering (ABE), part of the College of Agricultural, Consumer and Environmental Sciences and The Grainger College of Engineering at the U. of I. He is corresponding author on both studies.

“This work has many potential applications in the agricultural industry and can significantly lower costs. While a multispectral camera costs $10,000 or more, you can get an RGB camera for a few hundred dollars,” he added.

Analysis of sweet potato attributes
In the first paper, the researchers provide a large dataset of reconstructed images for chemical analysis of sweet potatoes that anyone can access and use for their own modeling.

“Most existing image reconstruction models focus on non-biological objects like tables and chairs, which are very different from biological objects. Our goal was to create an RGB-to-hyperspectral image dataset for a biological sample and make it publicly available,” said lead author Ocean Monjur, doctoral student in ABE.

Sweet potatoes are a popular food source, and they are also used for a wide range of industrial purposes including textiles, biodegradable polymers, and biofuels. Assessing quality attributes such as brix (sugar content), moisture, and dry matter is important for determining the usage and value of potatoes. Chemical laboratory analysis is time-consuming and destroys the samples. Hyperspectral imaging (HSI) is fast, accurate, and non-destructive, but it is expensive and complicated.

That’s why the researchers created Agro-HSR, a large database of reconstructed RGB to HSI images for the agricultural industry. The dataset includes 1322 image pairs from 790 sweet potato samples, collected from one or both sides of each potato. For 141 potato samples, they measured brix, firmness, and moisture content to evaluate the accuracy of the reconstructed images, finding them to be highly correlated with the actual measurements.

They tested their dataset on five popular hyperspectral imaging reconstruction models to determine which performed best, finding that two models (Restormer and MST++) consistently outperformed the others on all metrics.

“To our knowledge, this is the largest dataset for hyperspectral image reconstruction, not just for agriculture but overall. We are providing this database so anyone can use it to train or develop their own models, including models for other agricultural products,” Kamruzzaman said.

Evaluating chlorophyll content for maize growth
In the second paper, the researchers describe a novel method for multispectral image reconstruction to analyze chlorophyll content in maize. They also introduce a simple device that people can use to take pictures in the field and get immediate results.

“Our target measure is chlorophyll content, which is an indicator of plant growth. With this device you can take a picture, get the chlorophyll content, and determine the crop’s growth status,” Kamruzzaman said.

To develop their model, the researchers collected images from three different locations: a research field in Hengshui, China; the U. of I. Plant Biology Greenhouse; and the U. of I. Vegetable Crops Research Farm.

At each location, they divided the area into varying levels of soil fertility, and at the Illinois research farm, they subjected the maize to three levels of stress by flooding throughout the growth period.

In all of these settings, they tested several modeling approaches to reconstructing multispectral images from RGB. Based on their findings, they created a novel model called Window-Adaptive Spatial-Spectral Attention Transformer (WASSAT), which more accurately aligned with the actual data.

“We combined spectral and spatial attention modes to establish an adaptive window that can discern crops from soil and other elements, capturing the complexity of a field environment. Then we reconstructed 10-band images to predict chlorophyll content, and we found our results performed better than other models,” said lead author Di Song, doctoral student in ABE.

“We have developed a handheld device that incorporates the model. You can use it to take an RGB image, which will be converted to a multispectral image that provides much more information,” he said. “Next, we plan to add a prediction model, so the farmer can simply take a picture and get the chlorophyll content without having to interpret the images.”

This approach offers a cost-effective solution for accurate crop monitoring, enabling precise growth assessment and stress detection, the researchers concluded in the paper.