| | | | |  | | By Toni Odejimi, Erin Schumaker, Daniel Payne and Ruth Reader | | | | | 
AI's got a bias problem, but it might be fixable. | Justin Sullivan/Getty Images | Hospitals are collaborating with university health technology experts to form an alliance aimed at removing one of the biggest impediments to the adoption of advanced artificial intelligence tools in health care: bias. How so? The initiative, called VALID AI, began last year when two AI digital health specialists at the University of California, Davis, Dennis Chornenky and Ashish Atreja, convinced health care systems and research facilities to join their effort to establish industry standards for advanced AI. They hope to develop tools that collect more information about patients’ “social vital signs” — characteristics like socioeconomic status or access to care linked to health outcomes. One of the potential tools they’ve proposed is an AI toolkit of diverse data that better captures the role of social determinants of health. Armed with the toolkit, empowered providers could work more deliberately to improve health care outcomes by linking patients to community resources. VALID AI has more than 50 members, including New York-Presbyterian, Ochsner Health in Louisiana and Boston Children’s Hospital. Why it matters: Because artificial intelligence is based on data gathered by people, it reflects human prejudices against people of color, women and low-income patients. VALID AI believes it can reverse the problem by working with organizations that could train algorithms to detect bias and help patients. If the researchers succeed, they hope it will speed responsible adoption of AI tools to improve care, reduce disparities in access and diagnosis and make providers more efficient. “AI can analyze and synthesize vast amounts of health data incomprehensibly faster than a human or a bunch of humans could do to identify disparities in access and outcomes,” said Craig Kwiatkowski, chief information officer at one of the group’s founding members, Cedars-Sinai Medical Center in Los Angeles.
| | | | 
Copenhagen, Denmark | Shawn Zeller | This is where we explore the ideas and innovators shaping health care. Artificial intelligence analysis of aerial images helped U.S.-based researchers identify the location of fences in Wyoming, information needed to help wildlife like pronghorn migrate to escape harsh winters, Science reports. 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. | | | 
Uterus transplants are helping women have children. | Getty Images | Uterus transplants in 20 women, enabling 70 percent of them to have children, are offering hope to people with infertility, according to a new JAMA study. How’s that? The researchers looked at a small group of women, ages 20 to 36, without functioning uteruses to see whether transplanted uteruses from donors could allow them to conceive. All the transplants were done at Baylor University Medical Center in Dallas. Of the 20 patients, 14 delivered healthy children. However, complications such as blood clots prevented six women from bringing their babies to term. Uterus transplants are an emerging field of research for OB-GYNs. The first transplant was performed in 2011, and since then transplants have been offered across the world. Although no registry exists that precisely reports the number of live births resulting from these procedures, it is estimated that there are more than 70 live births worldwide. As of May 1, 2024, there have been 48 uterus transplants and 33 live births in the U.S. Next steps: Expanding the research on uterus transplants. Dr. Liza Johannesson, medical director of the uterus transplant program at Baylor and a study co-author, acknowledged the small trial size and lack of diversity in her study, with only three women of color participating. It’s also notable that the transplants were performed at a single center, and outcomes may differ in other settings. The next study will include more patients and more women of color. The researchers plan to publish results annually to track how patients and their children are doing. Johannesson is also pushing for insurance coverage of the procedure, which would make it more accessible. “I want that woman or that young girl who comes to that gynecologist to hear that uterus transplantation might be an option for her in the future, and right now, they’re not. They are left with a diagnosis with no treatment when they leave that doctor,” she said.
| | | | |  That frankfurter isn't the healthiest choice. | Getty Images | A diet heavy in red and processed meat is likely a main risk factor associated with younger adults developing colorectal cancers, Cleveland Clinic researchers say. In a new study published in the journal npj Precision Oncology researchers built an artificial intelligence algorithm to analyze data from Cleveland Clinic patients who’d gotten care for young- or average-onset colorectal cancer. The algorithm analyzed the data from patients’ microbiomes and metabolites, or the molecules the body produces when it breaks down food. They found that patients under 60 with colorectal cancer had a higher level of metabolites associated with arginine, an amino acid, and the urea cycle that clears nitrogen from the body, than the older patients did, a difference that might be tied to long-term consumption of red and processed meat. Why it matters: Incidence rates for more than a dozen cancer types, including colorectal cancers, are rising in younger people, according to a large study by researchers at the American Cancer Society and published in The Lancet Public Health this month. Even so: We can’t perform yearly colonoscopies on everyone under 60. “What is much more feasible is to give everyone in the system a simple test to measure a biomarker that determines their colorectal cancer risk,” Dr. Suneel Kamath, study co-author and gastrointestinal oncologist, said in a statement. “Then we can give the most at-risk individuals appropriate screening.” What’s next? Additional research is needed to determine whether the Cleveland Clinic results can be extrapolated more broadly. If the researchers can validate their results nationally, they want to test whether certain diets, or drugs that regulate arginine production and the urea cycle could prevent or treat colorectal cancers in young people. The takeaway: Talk to your doctor about your diet. Kamath says the findings have already changed the way he talks about diet risks: “I always make sure to bring it up to my patients.” | | | | Follow us on Twitter | | | | Follow us | | | | |
No comments:
Post a Comment