Microsoft has announced a new collaboration with Mass General Brigham, the University of Wisconsin School of Medicine and Public Health, and its affiliated health system, UW Health. The partnership aims to develop advanced artificial intelligence (AI) models focused on radiology and create innovative medical imaging applications.
The joint effort will explore how AI applications and algorithms can assist radiologists in interpreting medical images, generating reports, analyzing data, and classifying diseases. The partners plan to integrate these AI tools into clinical workflows, including Microsoft’s Nuance PowerScribe radiology reporting platform.
These new applications will be developed using Microsoft’s Azure AI platform and will build upon existing Nuance radiology applications.
Dr. Keith J. Dreyer, chief data science officer and chief imaging officer at Mass General Brigham, emphasized the transformative potential of generative AI. “Generative AI has the potential to break down traditional barriers in AI product development and speed up its impact on clinical care. We must develop and assess these tools responsibly to ensure they enhance, rather than compromise, the quality of care,” Dreyer said.
The foundation models, fine-tuned with Mass General Brigham’s extensive multimodal longitudinal data, are expected to shorten the development cycle for AI-based medical software. This includes automating the segmentation of organs and abnormalities in medical imaging, thereby boosting radiologists’ efficiency and consistency.
The Larger Context
Recently, Mass General Brigham faced significant disruptions due to a misconfigured CrowdStrike Falcon update. This issue caused a global IT outage on a Friday, although the health system continued to provide care for urgent cases. Non-urgent surgeries and medical visits were postponed during the outage.
This partnership is not the first collaboration between these institutions and Microsoft on AI in healthcare. At the HIMSS24 conference in March, Microsoft announced the Trustworthy & Responsible AI Network (TRAIN). This consortium, which includes Mass General Brigham and UW Health, aims to implement responsible AI practices to ensure safe and reliable use of AI in healthcare.
TRAIN comprises 16 health systems, including AdventHealth, Advocate Health, Boston Children’s Hospital, Cleveland Clinic, Duke Health, Johns Hopkins Medicine, MedStar Health, Mercy, Mount Sinai Health System, Northwestern Medicine, Providence, Sharp HealthCare, University of Texas Southwestern Medical Center, and Vanderbilt University Medical Center.
In a related development, Google Cloud and Bayer also announced a partnership in April. Their collaboration focuses on creating AI applications to reduce radiologist burnout and enhance diagnostic efficiency while ensuring data security.
