On Tuesday at the HIMSS conference in Orlando, GE HealthCare and Mass General Brigham announced that they are integrating foundation models into their AI research and development. The two partners began their AI research collaboration in 2017.
Since they launched their collaborative effort, the partners have “done a lot of work on operational efficiency,” Parminder Bhatia, GE HealthCare’s chief AI officer, said during a Wednesday interview at the conference.
With the Rise of AI, What IP Disputes in Healthcare Are Likely to Emerge?
Munck Wilson Mandala Partner Greg Howison shared his perspective on some of the legal ramifications around AI, IP, connected devices and the data they generate, in response to emailed questions.
For instance, he highlighted an algorithm they co-developed to help providers better identify missed care opportunities.
The algorithm helps providers predict which patients are likely to be late or not show up to appointments, and it has a 96% accuracy rate, Bhatia declared. By using the tool, healthcare providers can conduct more walk-in appointments, he explained.
The research partners have also developed several AI tools to help save radiologists time by flagging important findings in images, he noted.
By integrating foundation models into their AI research, GE Healthcare and Mass General Brigham are seeking to accelerate the development of more tools like this, Bhatia said.
A Personalized Approach to Medication Nonadherence
At the Abarca Forward conference earlier this year, George Van Antwerp, managing director at Deloitte, discussed how social determinants of health and a personalized member experience can improve medication adherence and health outcomes.
A foundation model refers to a large-scale machine learning model that serves as a basis for building more specialized models for specific tasks or domains. Typically, these models are pre-trained on large amounts of diverse data and are capable of understanding and generating human-like text with a high degree of accuracy.
When a health system deploys AI, it usually has to retrain models to suit the distinct needs of various patient groups and care settings. This often results in higher expenses and complexity, hurting the widespread acceptance of AI technologies within the healthcare sector, Bhatia noted.
“Foundation models really help accelerate the time-to-market for AI products,” he remarked. “Any future GE HealthCare AI developer who wants to build an application can now build the application about 10 times faster than in the past.”
The models can also help developers more quickly adapt existing AI tools for different use cases, Bhatia added.
Dr. Keith Dreyer, Mass General Brigham’s chief data science officer, called foundation models “the next wave of AI innovation” in a statement.
“I think we are all optimistic that foundation models may actually complement and enhance the work we have been doing with convolutional neural networks over the past few years. Hopefully, this work will help make healthcare delivery more efficient for our practitioners, more accessible for our patients and more equitable for our diverse communities,” he declared.
Photo: Andrzej Wojcicki, Getty Images
