Prof. Marzyeh Ghassemi speaks with NPR host Kate Wells about a decision by the National Eating Disorders Associations to replace their helpline with a chatbot. “I think it's very alienating to have an interactive system present you with irrelevant or what can feel like tangential information,” says Ghassemi.
Prof. Marzyeh Ghassemi speaks with WBUR reporter Geoff Brumfiel about her research studying the use of artificial intelligence in healthcare. “When you take state-of-the-art machine learning methods and systems and then evaluate them on different patient groups, they do not perform equally,” says Ghassemi.
A review led Prof. Marzyeh Ghassemi has found that a major issue in health-related machine learning models “is the relative scarcity of publicly available data sets in medicine,” reports Emily Sohn for Nature.
Principal research scientist Leo Anthony Celi speaks with Associated Press reporter Maddie Burakoff about how pulse oximeters can provide inaccurate readings in patients of color. Celi highlights how oxygen levels can also be measured by drawing blood out of an artery in the wrist, the “gold standard” for accuracy, but a method that is a a bit trickier and more painful.
Principal Research Scientist Leo Anthony Celi oversaw a study which found that people of color were given significantly less supplemental oxygen than white people because of inaccuracies in pulse oximeter readings, reports Nancy Lapid for Reuters. “Nurses and doctors make the wrong decisions and end up giving less oxygen to people of color because they are fooled [by incorrect readings from pulse oximeters],” says Celi.
A study co-authored by MIT researchers finds that algorithms based on clinical medical notes can predict the self-identified race of a patient, reports Katie Palmer for STAT. “We’re not ready for AI — no sector really is ready for AI — until they’ve figured out that the computers are learning things that they’re not supposed to learn,” says Principal Research Scientist Leo Anthony Celi.
Researchers at MIT and Harvard Medical School have created an artificial intelligence program that can accurately identify a patient’s race based off medical images, reports Tony Ho Tran for The Daily Beast. “The reason we decided to release this paper is to draw attention to the importance of evaluating, auditing, and regulating medical AI,” explains Principal Research Scientist Leo Anthony Celi.
An international team of scientists, including researchers from MIT and Harvard, have found that an artificial intelligence program trained to read X-rays and CT scans can successfully predict a person’s race with 90 percent accuracy, reports Hiawatha Bray for The Boston Globe. "The research effort was born when the scientists noticed that an AI program for examining chest X-rays was more likely to miss signs of illness in Black patients," writes Bray.
Isabella Cueto, a Cuban American journalist who has worked as a newspaper and radio reporter in Florida, South Carolina, and California, has been named the first recipient of the Sharon Begley-STAT Science Reporting Fellowship, reports STAT. “Named in honor of Begley, an award-winning science writer for STAT who died in January from complications of lung cancer, the fellowship combines a paid reporting position at STAT with an educational component provided through the prestigious Knight Science Journalism program.”
Prof. Evan Lieberman speaks with Craig LeMoult of GBH about his new study, which finds there are mixed reactions when people are informed of the racial disparities in Covid-19 outcomes in the U.S. “We are so interconnected as a society - economically, socially, politically,” says Lieberman, “and [it’s important] to remind everyone that we are all potential vectors for this epidemic so it really behooves all of us to cooperate and to be able to end this pandemic as soon as possible.”
Alyce Johnson, Interim Institute Community Equity Officer, and Sharon Bridburg, Director of HR for the Office of the Vice Chancellor, speak with Callie Crossley on WGBH’s "Under the Radar with Callie Crossley" about the importance of cross-racial friendships and their participation in The Club, a “diverse group of friends in the MIT and Harvard human resources community.”
In an article co-written for Smithsonian, Prof. John Van Reenen writes about an analysis he and his colleagues conducted examining how socioeconomic background, race and gender can impact a child’s chances of becoming an inventor. The researchers found that, “young people’s exposure to innovators may be an important way to reduce these disparities and increase the number of inventors.”
A study co-authored by MIT researchers finds African-American ridesharing passengers are more likely to experience cancellations and higher wait times, reports Chelsea Bailey for NBC News. "The patterns of discrimination were quite clear and consistent in both cities – and one can only assume it's happening all across the country," explains Prof. Christopher Knittel.
Prof. Christopher Knittel speaks with Lewis Wallace of Marketplace about his study that shows that Uber or Lyft passengers with “African-American sounding” names were more likely to be cancelled on or experience long wait times. “The chances that an Uber driver accepted their ride, observed their name, and then cancelled their ride more than doubled,” says Knittel.
A study co-authored by Prof. Christopher Knittel finds that some Uber and Lyft drivers are discriminating against passengers with “African-American sounding” names and taking women on longer, more expensive rides, writes Sara Ashley O’Brien for CNN Money. "We went into this hoping that we wouldn't see anything, but we found pretty strong evidence of discrimination," says Knittel.