Healthcare AI News – HIStalk – June 11, 2025
- The Joint Commission and the Coalition for Health AI are joining forces to create AI playbooks, tools, and a certification program, aiming to promote the responsible scaling of...
- A recent KPMG survey of 183 health system leaders across eight countries reveals that manny are struggling to operationalize AI use cases beyond the pilot stage.
- Early applications of artificial intelligence in healthcare include ambient documentation, image analysis, virtual health assistants for patient communication, early warning systems for patient deterioration, and claims and billing...
Healthcare AI is making waves, but many health systems still struggle to move beyond pilot programs, according to new findings.The challenges? Fragmented implementations and measuring return on investment. This HIStalk update reveals the Joint Commission’s bold move to create an AI certification program in collaboration with the Coalition for Health AI. It also delves into the top AI applications, including generative AI and machine learning, and the financial commitment—up to 10% of tech budgets—some healthcare providers are making to get ahead. News Directory 3 offers more on healthcare’s evolution. Learn about the FDA’s new AI tool, the latest advancements from OpenAI, and how companies like Mayo Clinic and Boehringer ingelheim are investing in the future.Discover what’s next for AI in healthcare now.
Joint Commission to Develop AI Certification for Healthcare
Updated June 11, 2025
The Joint Commission and the Coalition for Health AI are joining forces to create AI playbooks, tools, and a certification program, aiming to promote the responsible scaling of artificial intelligence in healthcare. This initiative comes as healthcare organizations navigate the complexities of implementing AI solutions.
A recent KPMG survey of 183 health system leaders across eight countries reveals that manny are struggling to operationalize AI use cases beyond the pilot stage. Common obstacles include fragmented implementations, difficulties in measuring return on investment, and workforce culture issues. The survey also found that 85% of respondents are self-developing AI solutions.
Early applications of artificial intelligence in healthcare include ambient documentation, image analysis, virtual health assistants for patient communication, early warning systems for patient deterioration, and claims and billing processing. The top five AI applications are generative AI, speech recognition, agentic AI, machine learning, and robotics. Notably, one-third of respondents reported that AI spending accounts for 10% or more of their technology budget.
In related news, OpenAI has released 03-pro, an AI model capable of performing PhD-level math and science tasks.The company has also considerably reduced the price of o3 by 80%.
Apple recently announced Apple Intelligence enhancements, including on-device live translation for Messages, FaceTime, and Phone, as well as contextual actions triggered by iPhone screen content. However, the companyS WWDC announcements did not include updates regarding the integration of AI into Siri.
The food and Drug Administration (FDA) has launched Elsa, an AI tool designed to summarize adverse events, compare product labels, and generate database code for non-clinical use. The agency reports that Elsa, deployed ahead of schedule and under budget, is already accelerating clinical protocol reviews and aiding scientific evaluations. However, some insiders have reported that the FDA’s AI tools are buggy and struggle with basic tasks.
Business Developments in AI and Healthcare
Mayo Clinic is investing in and collaborating with Hellocare.ai to develop ambient clinical intelligence technology. Hellocare.ai’s platform passively monitors clinical conversations, detecting care-related events to trigger documentation and workflow actions, according to CEO Labinot Bytyqi.
Boehringer Ingelheim’s animal health unit will integrate its canine heart murmur detection algorithms into Eko Health’s digital stethoscopes, enhancing diagnostic capabilities for veterinarians.
Clinical decision support developer openevidence has signed an agreement to incorporate data from 13 journals published by JAMA Network, expanding its resources for evidence-based recommendations.
AI Research Advancements
Researchers have developed an agentic AI system for selecting cancer treatments that aligns with expert conclusions 91% of the time. This system demonstrated improved decision-making accuracy compared to GPT-4, increasing from 30% to 87%, and correctly cited recognized oncology guidelines in 75% of its responses.
What’s next
as AI continues to evolve, healthcare organizations are focusing on responsible implementation, addressing challenges related to data privacy, accuracy, and integration with existing systems. the Joint Commission’s certification program represents a significant step toward ensuring the safe and effective use of AI in delivering better healthcare.