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.”
Forbes contributor Adi Gaskell writes that CSAIL researchers have developed a machine learning system that can determine whether a task is best performed by a human or AI. The researchers developed the system to be “capable of learning and adapting as it goes, such that it can identify,” Gaskell explains, “when the expert isn't available or whether they have a certain level of experience, before choosing whether to defer to them.”
Forbes reporter Eva Amsen writes about a new study by researchers from the Media Lab that explores how to credit art developed by AI systems. The researches found that “credit for AI-generated art all depends on how we think and talk about the role of AI.”
CNBC reporter Charlie Wood features tProf. Connor Coley's work developing a new system that could be used to help automate molecule manufacturing. “It tries to understand, based on those patterns, what kind of transformations should work for new molecules it’s never seen before,” says Coley.
Axios reporter Erica Pandey writes that a new working paper by MIT researchers explores the potential pitfalls posed by using algorithms to aid the hiring process. "Lots of companies have taken interest in using AI tools in the recruiting process," explains Prof. Danielle Li. "In that world, algorithms stand to have a big impact."
Prof. Tonio Buonassisi writes for Indvstrvs about how businesses can use machine learning systems to help address some of the world’s most pressing challenges. “What has emerged through our research,” writes Buonassisi, “is a combination of automation, data science, and computation — building blocks that promise to accelerate the rate of new materials development tenfold, and eventually millions of times, faster.”
NBC News reporters Lindsay Hoffman and Caroline Kim spotlight graduate student Joy Buolamwini’s work uncovering racial and gender bias in AI systems in a piece highlighting women who are “shattering ceilings, making groundbreaking discoveries, and spreading public awareness during the global pandemic.” Hoffman and Kim note that Buolamwini’s research "helped persuade these companies to put a hold on facial recognition technology until federal regulations were passed.”
Writing for The Boston Globe, Prof. D. Fox Harrell, Francesca Panetta and Pakinam Amer of the MIT Center for Advanced Virtuality explore the potential dangers posed by deepfake videos. “Combatting misinformation in the media requires a shared commitment to human rights and dignity — a precondition for addressing many social ills, malevolent deepfakes included,” they write.
Third-year student Casey Johnson speaks with Washington Post reporter Luz Lazo about his work exploring the feasibility of using GPS technology to determine when a scooter is on the sidewalk. Lazo explains that Johnson wrote a “surface categorization algorithm to detect the periodic cracks in a sidewalk. He then added an accelerator sensor — which costs less than $1 — to detect when the scooter is being used on an asphalt road versus a concrete sidewalk.”
Researchers from the Singapore-MIT Alliance for Research and Technology (SMART) Critical Analytics for Manufacturing Personalized Medicine (CAMP) research group have been awarded new research grants aimed at supporting work exploring personalized medicine and cell therapy, reports Health Europa. “In addition to our existing research on our three flagship projects, we hope to develop breakthroughs in manufacturing other cell therapy platforms that will enable better medical treatments and outcomes for society,” says Associate Provost Krystyn Van Vliet.