| | | | |  | | By Ruth Reader, Daniel Payne, Erin Schumaker and Toni Odejimi | | | | | 
A health tech firm is making use of data about what the doctor doesn't do. | AP Photo/Mike Derer | Artificial intelligence needs lots of data to make its predictions, and missing data can result in less accurate conclusions. But newly FDA-cleared artificial intelligence software eCART, which predicts when patients might take a turn for the worse, makes use of missing data. “Missingness is actually not a bug, it’s a feature,” said Dr. Dana Edelson, founder of eCART maker AgileMD. Her early warning tool uses routine patient data, like vital signs and routine lab work. But the algorithm is trained to understand that missing data can gives clues to what the doctor is thinking. For example, if a patient hasn’t received a test for a certain digestive enzyme called lipase, it’s probably because the doctor doesn’t think the patient has pancreatitis, Edelson said. It seems to work: eCart showed that it could reduce in-hospital mortality in a peer-reviewed study. And in a yet-to-be reviewed preprint study, it performed better than the U.K.’s National Early Warning System, which detects the severity of patient illness. Why it matters: Missing patient data can affect how well health artificial intelligence works. The FDA seems to be paying attention to how well companies deal with this problem. Health tech company Prenosis, maker of the only algorithm to receive FDA authorization for predicting sepsis, a sometimes fatal immune response to infection, doesn’t deliver a decision unless it has all the required data. It also shows each data point influencing its predictions.
| | | | Understand 2024’s big impacts with Pro’s extensive Campaign Races Dashboard, exclusive insights, and key coverage of federal- and state-level debates. Focus on policy. Learn more. | | | | | | | | | 
Lake George, N.Y. | Erin Schumaker/POLITICO | This is where we explore the ideas and innovators shaping health care. The National Institutes of Health is launching a center today to advance innovation in taste and smell research, Director Monica Bertagnolli announced on X. Share any thoughts, news, tips and feedback with Carmen Paun at cpaun@politico.com, Daniel Payne at dpayne@politico.com, Ruth Reader at rreader@politico.com, Erin Schumaker at eschumaker@politico.com, or Toni Odejimi at aodejimi@politico.com. Send tips securely through SecureDrop, Signal, Telegram or WhatsApp.
| | |  Private equity owns more of these machines than ever. | Getty Images | Private equity ownership of cardiology clinics in the U.S. has accelerated significantly since 2021, according to an early draft of a report set to be published in the Journal of the American College of Cardiology, in which researchers called for close monitoring of care after changes in ownership. Of the 342 clinics bought by private equity groups, 94 percent were acquired from 2021-2023, the report found. Sixty-four of them were acquired twice through the study period, from 2013-2023. Private equity groups focus their efforts in wealthier communities and tend to favor relatively short-term investments over three to seven years, the researchers found. Even so, relatively few cardiology clinics — about 3 percent — were acquired by private equity groups. Meanwhile, in Washington: Some lawmakers have worked to increase scrutiny of how private equity ownership impacts care across the health system. Sen. Ed Markey (D-Mass.), for example, offered draft legislation that would hold private equity firms accountable when they buy health providers. Why it matters: Some researchers and advocacy groups have suggested that private equity leads to higher costs and lower quality, a claim private equity companies deny. Providers, policymakers and patients have all grown increasingly interested in the role of financiers in care delivery.
| | | | Understand 2024’s big impacts with Pro’s extensive Campaign Races Dashboard, exclusive insights, and key coverage of federal- and state-level debates. Focus on policy. Learn more. | | | | | | | | | 
ARPA-H wants to bring down the cost of an MRI. | Keith Srakocic/AP Photo | The high-risk, high-reward Advanced Research Projects Agency for Health is backing a plan to develop a cheaper magnetic resonance imaging device for breast cancer detection. Its $22 million award to the Yale School of Medicine aims to fund the development of an MRI that’s 10 times cheaper than current MRIs. Why it matters: Breast cancer is the second most common cancer among women, according to the Centers for Disease Control and Prevention, with more than 270,000 new cases reported each year. Early detection improves women’s chance of survival, and MRIs are twice as sensitive at detecting breast cancer than mammograms, the standard screening method. But MRIs are more expensive than mammograms and aren’t readily available, except in large radiology centers. “MRI is by far the best imaging modality for breast cancer detection,” Todd Constable, the project’s principal investigator and a radiology and biomedical imaging professor at Yale School of Medicine, said in a statement. What’s next? Constable’s team is working to create a device with a smaller footprint and cheaper magnets, which would allow it to be used more widely. “This will allow these devices to be placed in community health clinics, greatly expand women’s access to MRI screening, and allow such screening to be part of a women’s annual checkup,” Constable said. | | | | Follow us on Twitter | | | | Follow us | | | | |
No comments:
Post a Comment