Health

Research scientist Brian Subirana speaks with The Economist’s Babbage podcast about his work developing a new AI system that could be used to help diagnose people asymptomatic Covid-19.

Prof. James Collins speaks with Boston 25 reporter Julianne Lima about the growing issue of antibiotic resistant bacteria and his work using AI to identify new antibiotics. Collins explains that a new platform he developed with Prof. Regina Barzilay uncovered “a host of new antibiotics including one that we call halicin that has remarkable activity against multi drug-resistant pathogens.”

A new algorithm developed by MIT researchers could be used to help detect people with Covid-19 by listening to the sound of their coughs, reports Zoe Kleinman for BBC News. “In tests, it achieved a 98.5% success rate among people who had received an official positive coronavirus test result, rising to 100% in those who had no other symptoms,” writes Kleinman.

Mashable reporter Rachel Kraus writes that a new system developed by MIT researchers could be used to help identify patients with Covid-19. Kraus writes that the algorithm can “differentiate the forced coughs of asymptomatic people who have Covid from those of healthy people.”

A new took developed by MIT researchers uses neural networks to help identify Covid-19, reports Alyse Stanley for Gizmodo. The model “can detect the subtle changes in a person’s cough that indicate whether they’re infected, even if they don’t have any other symptoms,” Stanley explains.

TechCrunch reporter Devin Coldewey writes that MIT researchers have built a new AI model that can help detect Covid-19 by listening to the sound of a person’s cough. “The tool is detecting features that allow it to discriminate the subjects that have COVID from the ones that don’t,” explains Brian Subirana, a research scientist in MIT’s Auto-ID Laboratory.

MIT researchers have developed a new AI model that could help identify people with asymptomatic Covid-19 based on the sound of their cough, reports CBS Boston. The researchers hope that in the future the model could be used to help create an app that serves as a “noninvasive prescreening tool to figure out who is likely to have the coronavirus.”

Tech Explorist reporter Amit Malewar writes that researchers from Singapore-MIT Alliance for Research and Technology (SMART) have “demonstrated a new way to manufacture human red blood cells (RBCs) that cuts the culture time by half compared to existing methods.”

New research from Prof. Lydia Bourouiba finds that guidelines for safe social distancing may need to be updated as researchers gain more information about how the virus spreads, reports Ben Guarino for The Washington Post. “It becomes very important to not think just about a fixed distance. It’s very important to think about the air flow,” says Bourouiba. 

Researchers from MIT and the University of Oxford have found that guidelines for social distancing during the Covid-19 pandemic may need to be updated to account for factors such as ventilation, crowd size and exposure time, reports Akshay Syal for NBC News.  "It's not just 6 feet and then everything else can be ignored or just mask and everything else can be ignored or just ventilation and everything else can be ignored,” says Prof. Lydia Bourouiba.

MIT researchers have created a new synthetic lining that could be used to help treat digestive diseases, reports Pratibha Gopalakrishna for STAT. “We wanted to develop a liquid system that is easier to take compared to tablets or capsules, but had enhanced capabilities,” explains Prof. Giovanni Traverso.

New Scientist reporter Alice Klein writes that MIT researchers have developed a synthetic glue that could be used to line the small intestine and help treat several conditions, including diabetes and lactose intolerance. “The researchers found they could control the uptake of different nutrients,” writes Klein, “by adding various substances to the synthetic lining.”

A new study co-authored by MIT researchers finds that Covid-19 may shut down germinal centers, which help patients mount an antibody response. The findings “may help explain COVID-19 progression in the sickest cases and provide important insights to vaccine developers,” reports Jon Cohen for Science.

Newsweek reporter Meghan Roos writes that a study co-authored by Prof. Victor Chernozhukov finds a nationwide mask mandate in the U.S. could have reduced Covid-19 deaths by 40% among workers who regularly interact with the public. The researchers found “a national mask mandate for workers could have prevented between 17,000 and 55,000 deaths between the start of April and the start of June.”

Prof. James Collins speaks with Molly Wood of Marketplace about his work developing a faster, cheaper and more accurate Covid-19 diagnostic. Collins explains that his research group is “using synthetic biology to create highly sensitive, low-cost diagnostics, some that are now approved for use in clinical diagnostics labs, and now we’re moving towards point-of-care diagnostics, as well as at-home diagnostics.”