Artificial intelligence (AI) and machine-learning (ML) driven precision medicine company Zephyr AI announced on Thursday a partnership with MedStar Health that will leverage real-world health data to develop predictive models of chronic diseases. The initial focus of the partnership will be on type 2 diabetes (T2D), using MedStar’s de-identified T2D data set.
“We are laser-focused on delivering results that improve the timeliness of T2D interventions and decrease devastating adverse health outcomes commonly associated with the chronic condition, such as end-stage chronic kidney disease, dialysis, foot ulcers, and amputations,” said Grant Verstandig co-founder and executive chairman, and CEO of Zephyr AI and also a founding partner at venture capital firm Red Cell Partners.
MedStar Health is a $6.7 billion integrated health network operating more than 300 locations in Virginia, Maryland, and Washington, D.C. In addition to providing healthcare in the region, MedStar is also very active a research and educational entity. Founded in 1963 the institute serves more than 1,000 associates across 40 locations researching a broad range of diseases and also has an active clinical trials arm.
“Our partnership with Zephyr AI is a breakthrough opportunity to develop innovative artificial intelligence and machine learning capabilities, starting with type 2 diabetes, to benefit our patients and our providers,” said Stephen R.T. Evans, MD, MedStar Health’s executive vice president and CMO. “Together, Zephyr AI and MedStar Health can develop technology that brings the promise of precision medicine to chronic diseases and that complements our existing programs to improve outcomes and reduce costs for people with diabetes.”
Zephyr AI, founded in 2021 via the incubation firm Red Cell Partners, is leveraging data and an AI/ML approach to solving press healthcare problems. It largely operates in three areas: oncology, where its Vulnerability Network Method, helps develop algorithms to classify tumors while revealing new targets for drug discovery, label expansion, and precision care; predictive analytics for population health, patient management and clinical decisions support; and data as a service which standardizes disparate data sets to enable more effective research.
For the MedStar partnership, Zephyr AI will apply its Insights clinical decision support tool to help refine T2D predictive models and inform patient care.
“While there have always been well-known population risk factors associated with adverse outcomes in patients diagnosed with T2D, this partnership will allow Zephyr AI to train our revolutionary machine learning models on one of the most robust T2D data sets in the world,” said David L. Morgan II, CEO of Zephyr AI. “The combination of our team’s world-class AI with MedStar Health’s data and expertise dramatically increases our collective prediction accuracy and will help both partners shift from a population-level predictive capability to an individual-level predictive capability. This relationship means that we can leverage real-world data from a renowned health system that has one of the most advanced, innovative, and successful T2D programs in the U.S. Ultimately, we believe we can help patients better manage their diabetes by developing predictive tools for clinicians.”