Mount Sinai Secures $4 Million for Kidney Injury AI

The National Institutes of Health awarded a five-year grant to develop a surgical predictive model.

Updated on Oct. 27, 2026 in Artificial Intelligence

Isometric editorial illustration of a glass beaker filled with glowing digital lattice prisms, representing medical AI research. — The National Institutes of Health awarded a $4 million grant to Mount Sinai to develop an AI model for predicting kidney injury. AI Illustration. Upload story photo >

Live Poll

Do you believe using artificial intelligence to analyze patient health records will improve surgical outcomes?

The National Institutes of Health awarded a $4 million grant to the Icahn School of Medicine at Mount Sinai to build an artificial intelligence model for predicting postoperative acute kidney injury. The project aims to improve outcomes for patients following major noncardiac surgeries.

Why it matters

Current methods for predicting kidney injury are unreliable and often based on small patient samples. This research seeks to create a more robust prediction tool to inform future personalized prevention strategies.

The study utilizes federated learning to analyze data from over 320,000 patients across 15 hospitals without transferring individual patient records. Researchers will evaluate this against models built using pooled data to assess performance efficacy.

The players

Icahn School of Medicine at Mount Sinai

This New York City-based institution is the lead research body for the kidney injury AI project.

National Institutes of Health

This federal agency is the primary government funding body for medical research in the United States.

Vanderbilt University

This academic research university is one of the 15 participating institutions providing patient data.

University of California, Los Angeles

This public research university is participating as a partner institution in the multi-hospital study.

The details

The research team will analyze data from 15 institutions, including Vanderbilt University and the University of California, Los Angeles, to refine the predictive algorithm. This approach targets the 12 percent of the 50 million annual U.S. noncardiac surgeries that are currently complicated by acute kidney injury.

October 27, 2026: The research grant award was officially announced.
Five-year period: The expected duration of the National Institutes of Health research project.

The Tech Race

This project represents a broader transition in healthcare toward federated learning, which enables AI development while preserving patient data privacy. It replaces traditional, siloed data pooling methods that have historically hindered the development of robust, large-scale predictive models.

While this tool is currently in development, it could eventually lead to safer surgical outcomes by identifying at-risk patients before complications arise. Patients undergoing noncardiac procedures may benefit from more personalized preventative care based on these future findings.

The takeaway

This project highlights how hospitals are leveraging federated learning to overcome data-sharing hurdles in medical research. By analyzing larger datasets, researchers hope to create more reliable diagnostic tools for common surgical complications.