Artificial intelligence, or the use of computing to simulate human intelligence, seems to be becoming more mainstream in healthcare applications. Many hospitals and healthcare facilities have adopted or are evaluating at least one AI program, including sifting through thousands of X-ray images to identify abnormalities and helping with staff planning.
Even if you don’t cover health informatics consistently, you may find yourself writing about AI at some point. Therefore, I collected a new tip sheet which covers some basics to help people get up to speed or gain a basic understanding of the subject. AI will also be the subject of a session at the AHCJ Health Journalism 2023 conference in Saint Louis.
I’ve seen a slew of stories about AI in healthcare this fall. Here is an overview of some interesting news or trends that journalists can use as starting points for their own stories:
- President Biden unveiled in October an AI bill of rights, which outlines five protections Americans should enjoy in an age of increasing use of technology, data, and automated systems. Specifically, the document states that people should be protected from unsafe or inefficient systems, and that automated systems should be developed in consultation with diverse communities and should be tested for safety. People should not be discriminated against by algorithms and systems should be designed fairly. People need to be protected against abusive data practices and be able to decide how data about them is used. They must be informed when automated systems are used and understand how and why these systems contribute to the results that affect them. Finally, they need to be able to opt out of systems when needed and have access to someone who can remedy any issues quickly.
- The FDA rapidly approved medical AI tools, with 521 devices approved, a 65% increase over the past two years, according to articles in Medtech Diving and the December 20 Healthtech e-newsletter from STAT. Some 91 devices were cleared by the FDA in 2022 alone, including one AI-powered solution for heart failure developed at the Mayo Clinic called EchoGo, which uses AI to detect heart failure from a single electrocardiogram image. Three-quarters of approved devices were in radiology and 11% in cardiology. How many hospitals are actually using the products and how they are improving patient outcomes will be the subject of an ongoing series on AI in medicine that STAT has scheduled for 2023, the e-newsletter says.
- A group of university hospitals, government agencies and private companies have come together to form the Coalition for Healthcare AI (CHAI) to foster high-quality healthcare by promoting the adoption of credible, fair and transparent healthcare AI systems. Coalition members include Change Healthcare, Duke AI Health, Google, Johns Hopkins University, Mayo Clinic, Microsoft, MITER, Stanford Medicine, UC Berkeley, and UC San Francisco. “The application of AI brings enormous benefit to patient care, but so does its potential to exacerbate inequalities in healthcare,” said coalition co-founder John Halamka, MD, president of Mayo Clinic Platform, in a statement on the CHAI website. “Guidelines for the ethical use of an AI solution cannot be an afterthought. Our coalition’s experts share their commitment to ensuring that patient-centered, stakeholder-informed guidelines achieve equitable outcomes for all populations.
- The NIH launched in September the Bridge to Artificial Intelligence (Bridge2AI) program to accelerate the widespread use of AI by the biomedical and behavioral research communities. The program brought together team members from diverse disciplines and backgrounds to generate datasets and best practices to prepare biomedical and behavioral research data for future AI applications. The agency awarded four prizes for data-generating projects in areas such as the use of imaging and other clinical data collected in an intensive care unit for diagnosis and risk prediction and using voice as a biomarker of human health.
- Many medical centers have also announced many exciting ways to integrate or test AI. These include:
- Google Cloud and Hackensack Meridian Health in New Jersey started collaborating using Medical Imaging Suite software trying to improve cancer detection, reported Becker’s Hospital Review. The healthcare system uses the program to anonymize imaging data and create AI algorithms to predict metastasis in prostate cancer patients.
- Using machine learning, Mayo Clinic researchers are studying patterns of change in pregnant patients in labor to help determine if vaginal birth can occur with good outcomes for the patient and baby, according to a story in Healthcare IT News. The models could predict the likelihood of cesarean delivery, postpartum hemorrhage, or amniotic infections, among other scenarios.
- Researchers at the Massachusetts Institute of Technology studied how AI could diagnose Parkinson’s disease by analyzing respiratory data. His ability to do so was published in a study in natural medicine. The system, trained on data collected during nearly 12,000 nights of sleep, was able to accurately flag patients with Parkinson’s disease and differentiate them from patients with Alzheimer’s disease, according to STAT’s e-newsletter on health technologies. Researchers are working on a follow-up study to determine whether AI can detect Parkinson’s disease earlier than conventional methods that rely on subjective expert assessments.
- At Beaumont Hospital in Michigan, researchers have started using an AI-powered blood test to detect fetal heart defects, Becker’s hospital review reported.
Here are some other resources to inform your reports. Happy Holidays!