GoKawiilLooking at a childlike version of your face might help you see childhood memories more clearly. Plus, some AI disease-prediction models are trained on suspect data and how thermal imaging can reveal night-flying birds.
Hello Nature readers, would you like to get this Briefing in your inbox free every day? Sign up here.
To capture images of birds, such as this yellow-rumped warbler (Setophaga coronata), flying at night, one research-team member detected the thermal signature of the bird and shone a torch at it. Another team member then took a snap with a digital camera. (Ross Gallardy/Macaulay Library)
Researchers have used thermal imaging technology to reveal and track the movements of bird species that travel at night. Some birds complete legs of their migrations at night because the air is cooler than during the day, and possibly to make it more difficult for predators to see them. But their nocturnal activities have long stumped ornithologists wanting to catalogue the birds’ movements. The ability to visualize night-flying birds could help reveal which species are most vulnerable to threats such as wind turbines and light pollution, says ecologist Felix Liechti.
Science | 6 min read
Reference: Ornithology paper
Dozens of artificial-intelligence models designed to predict a person’s risk of stroke or diabetes are trained on dubious datasets. Researchers found that many such models used one of two open-access health datasets that provide little information about where the data come from, and contain multiple oddities that suggest the data could have been fabricated. At least two of these models have been used in hospitals in Indonesia and Spain, although it’s unclear whether this has led to any flawed diagnoses.
Nature | 7 min read Reference: medRxiv preprint (not peer reviewed)
The US National Science Foundation (NSF) has announced a record number — 2,599 — of prestigious Graduate Research Fellowship Program awards. The increase came as a surprise after the number of awards was cut in half last year, this year’s application window opened late and some submissions were rejected without peer review. Quantum science and AI — priorities of the administration of US President Donald Trump — were the big winners this year, but biological sciences also got an unexpected boost.
Nature | 7 min read
... continue reading