AI Diagnoses Deadly Disease; Radiology’s New AI Guidance; WHO Tackles LMMs

Welcome to MedAI Roundup, highlighting the latest news and research in healthcare-related artificial intelligence each month.

Researchers at Children’s National Hospital in Washington, D.C., have developed an artificial intelligence (AI)-powered clinical tool for diagnosing rheumatic heart disease — which kills about 400,000 people a year worldwide — early enough to treat patients with penicillin and avoid the need for surgery. (Washington Post)

Five radiology societies published a joint statement on the specialty’s use and development of AI tools in the Journal of the American College of Radiology. Among the recommendations are increased safety monitoring; greater collaboration among developers, clinicians, purchasers, and regulators; and more focus on responsible and ethical integration of the technology into practice.

Companies have started using AI to improve health insurance shopping by predicting a person’s healthcare needs, which experts said could be especially beneficial for people purchasing Medicare Advantage plans or using Affordable Care Act marketplaces. (Axios)

The WHO released new guidance on the ethics and governance of large multimodal models, or LMMs. Distinct from large language models (LLMs), these are AI models that can process images and videos as well as text. The guidance outlined 40 recommendations based on five types of applications, including clinical care, patient-guided use, administrative tasks, medical education, and scientific research and drug development.

The American Academy of Family Physicians in partnership with Navina, a healthcare AI company developing administrative support tools, released a report showing that AI-powered assistants can help healthcare providers reduce time spent preparing for patient visits by 38% and reduce self-reported burnout by 23%.

Clinical researchers from the University of Michigan Medicine have developed an AI algorithm that predicts in-hospital mortality and other major complications following percutaneous coronary intervention, which could be used as a clinical decision-making tool for treatment of blocked arteries, according to a study published in the European Heart Journal.

Scientists at the University of California San Diego developed a new AI algorithm that can predict when cancer will be resistant to chemotherapy, according to a study published in Cancer Discovery.

The healthcare AI company Paige, which specializes in digital pathology tools, announced its latest application can detect cancer across more than 17 different tissue types, including skin, lung, and the gastrointestinal tract.

Atul Butte, MD, PhD, of the Baker Computational Health Sciences Institute at the University of California San Francisco, explained how healthcare professionals can create “scalable privilege” by using AI to translate clinical research and data into better patient care in the latest edition of JAMA‘s interview series with editor-in-chief Kirsten Bibbins-Domingo, MD, PhD.

Please enable JavaScript to view the

comments powered by Disqus.