In December 1994, a weight-loss movement was started, stating that obesity(肥胖) had become America’s second-largest cause of death, “resulting in about 300, 000 lives lost each year.” This marked the beginning of a long and influential life for the number. Experts began using the number to stress the seriousness of the problem. After ten years, the number had ballooned to nearly 400, 000 deaths caused by high weight, and it was expected to soon reach half a million. But these seemingly shocking numbers were then overturned(推翻) The number killer was Katherine Flegal.

Hoping to better understand the relationship between body size and death rate, Flegal led a study that finally showed something many scientists—and certainly the public—did not expect:the annual(每年的) deaths related to obesity added up to just more than 100, 000. More interesting still, people considered “overweight” could live longer than those who had “normal” weight. The study was met with a firestorm of doubt. “People were saying that our article was rubbish,” Flegal says. “Our findings produced controversy(争论) because someone wanted a controversy.”

“People are just more comfortable with a larger number—they prefer it to be larger,” Flegal says. And in nutrition science and medicine, that larger number is equal to the idea that thinner is better, skinnier is healthier. “You can report the same findings about obesity 100 times, and it can be called ‘unexpected’ every single time,” Flegal says. “At what point would you change your mind and say this might be the expected?”

“It’s human nature for people to find proof that supports their belief,” Caan, a breast cancer researcher says. “But we need to study where the data takes us.” Her own “unexpected findings” led to a discovery that could influence breast cancer patients:keeping muscle during treatment is far more important to survival than avoiding weight gain. “We can’t be stuck in what we thought was true,” Caan says, “because then we’re never going to move the field forward.”

Human’s appetite for sand could increase 45 percent within four decades, according to researchers who say unchecked consumption risks environmental damage and shortages of a key material for urban expansion. Growing demand for building sand — which is used to make concrete, glass and other vital construction materials – has already seen the rise of sand pirates (盗贼) , with dozens of islands disappearing in Indonesia as a result of casual mining.

Xiaoyang Zhong at Leiden University and his colleagues have calculated that global building sand demand will jump from 3.2 billion tonnes a year in 2020 to 4.6 billion tonnes by 2060. The figure is based on a central situation of future population rises and economic growth, and modelled using estimates of concrete and glass consumption, and the floor area needed in buildings.But there is no reliable estimate for remaining sand reserves, so it is unclear if the world can bear such a big increase. “Sand, and the sand crisis (危机), has been overlooked, creating severe environmental and social consequences. If we don’t act now, we may not have enough sand to develop our cities,” says Zhong.

However, Zhong’s team found that about half the projected consumption in 2060 could be avoided if countries take a suit of measures, including extending the lifetime of buildings, reusing concrete, creating more lightweight building designs and using other materials, such as wooden frames. According to the model, the single biggest reduction in sand use could come from more efficient (有效的) use of space: distributing less floor space per person in buildings, sharing offices, and so on. “It’s hard to say how realistic these measures are. But we want this to happen,” says Zhong.

The research only looked at sand used for glass and concrete in buildings, so is an underestimate of total future demand. Granular data on sand consumption for the 26 world regions studied is also lacking, and not detailed enough for country-level breakdowns.

Failure to act will add existing environmental pressures on reserves of sand in lakes and rivers first, but absolute shortages shouldn't be overlooked, says Zhong. “It would be very questionable if this growing demand could be met,” he says.

Voices offer lots of information. It turns out that they can even help diagnose (诊断) an illness and researchers are working on an app for that. The National Institutes of Health is funding a massive research project to collect voice data and develop an AI that could diagnose people based on their speech.

Everything such as your breathing patterns when you speak offers potential information about your health, says Dr. Yael Bensoussan, the director of the University of South Florida’s Health Voice Center and a leader on the study.“We asked experts: Well, if you close your eyes when a patient comes in, just by listening to their voice, can you have an idea of the diagnosis they have?” says Bensoussan.“And that’s where we got all our information. Someone who speaks low and slowly might have Parkinson’s disease. Depression or cancer could even be diagnosed.”

The project is part of the NIH’s Bridge to AI program, which was launched over a year ago with more than $100 million in funding from the government, with the goal of creating large-scale health care databases for precision (精准) medicine.“We were really lacking what we call open source databases,” says Bensoussan.“Every institution has their own database. But to create these networks was really important to allow researchers from other generations to use this data.”

The ultimate goal of the project is an app that could help bridge access to rural or underserved communities, by helping general practitioners (行医者) refer patients to specialists. To get there, researchers have to start by amassing data, since the AI can only get as good as the database it’s learning from. By the end of the four years, they hope to collect about 30,000 voices.

There are a few roadblocks, however. HIPAA, the law that regulates medical privacy, isn’t really clear on whether researchers can share voices. Every institution has different rules on what can be shared, and that opens all sorts of moral and legal questions.