AI Model GLUCOSE Enhances Blood Sugar Management in Post-Cardiac Surgery Care

AI-driven model supports safer and more precise blood sugar management after heart surgery

Researchers at the Icahn School of Medicine at Mount Sinai have developed a machine learning tool that can help doctors manage blood sugar levels in patients recovering from heart surgery, a critical but often difficult task in the intensive care unit (ICU). The findings appear in npj Digital Medicine. After cardiac surgery, patients are at risk of both high and low blood sugar, which can lead to serious complications. Managing these fluctuations requires careful insulin dosing, but existing protocols often fall short due to the unpredictable nature of ICU care and differences among patients, say the investigators. To address this challenge, the research team created a reinforcement learning model, named GLUCOSE, that recommends insulin doses tailored to each patient's needs. In tests using data from real-world ICU cases, GLUCOSE matched or even outperformed experienced clinicians in keeping blood sugar levels within a safe range -- despite having access to only current patient data, while doctors used full patient histories. "Our study shows that artificial intelligence can be thoughtfully and responsibly developed to support, rather than replace, the clinical judgment of health care professionals," says co-senior corresponding author Ankit Sakhuja, MBBS, MS, Associate Professor of Medicine (Data-Driven and Digital Medicine) and a member of the Institute for Critical Care Medicine at the Icahn School of Medicine at Mount Sinai. "In complex and high-pressure environments like the ICU, tools like GLUCOSE can provide real-time data-driven guidance tailored to individual patients. This kind of decision support can enhance safety, reduce the risk of complications, and ultimately allow clinicians to focus more of their attention on critical aspects of patient care." The research team trained GLUCOSE using reinforcement learning, which allowed the system to make optimal decisions through trial and error. They also used advanced methods -- conservative and distributional reinforcement learning -- to ensure the model made cautious, reliable recommendations. The model was then rigorously evaluated and compared to real-world clinical practices. While the results are promising, the researchers caution that GLUCOSE is not intended to replace doctors. It serves as a clinical decision support tool, offering suggestions that physicians can choose to follow based on their judgment and the broader clinical picture. The model could eventually be integrated into electronic health record systems to provide real-time insulin dosing guidance in the ICU, helping reduce complications and improve outcomes. Future steps include adapting the tool for use in other hospital settings, running clinical trials, and exploring ways to integrate it into routine care. One current limitation is that the model does not yet factor in nutrition data, which may affect longer-term glucose control. Still, the ability of GLUCOSE to make accurate recommendations based on limited real-time data highlights its potential to enhance safety and efficiency in postsurgical care. "Our goal is to develop AI systems that meaningfully augment the capabilities of health care providers and ultimately improve patient outcomes," says co-senior corresponding author Girish N. Nadkarni, MD, MPH, Chair of the Windreich Department of Artificial Intelligence and Human Health, Director of the Hasso Plattner Institute for Digital Health, and Irene and Dr. Arthur M. Fishberg Professor of Medicine at the Icahn School of Medicine at Mount Sinai, and Chief AI Officer of the Mount Sinai Health System. "By learning from real-world clinical data and delivering personalized recommendations in real time, models like GLUCOSE represent an important advance toward integrating trustworthy data-driven tools into the clinical workflow. This study offers a glimpse of how AI can be thoughtfully embedded into care to support providers in delivering safer, more precise treatment."

AI-Driven Model Supports Safer and More Precise Blood Sugar Management After Heart Surgery | Newswise

Representative cases showing how GLUCOSE's insulin dosing compares to actual clinician decisions in internal (a-c) and external (d-f) testing. Solid lines show glucose levels; dashed lines show insulin doses. Colored bands mark glucose ranges. Newswise -- New York, NY [May 28, 2025] -- Researchers at the Icahn School of Medicine at Mount Sinai have developed a machine learning tool that can help doctors manage blood sugar levels in patients recovering from heart surgery, a critical but often difficult task in the intensive care unit (ICU). The findings were reported in the May 27 online issue of NPJ Digital Medicine. After cardiac surgery, patients are at risk for both high and low blood sugar, which can lead to serious complications. Managing these fluctuations requires careful insulin dosing, but existing protocols often fall short due to the unpredictable nature of ICU care and differences among patients, say the investigators. To address this challenge, the research team created a reinforcement learning model, named GLUCOSE, that recommends insulin doses tailored to each patient's needs. In tests using data from real-world ICU cases, GLUCOSE matched or even outperformed experienced clinicians in keeping blood sugar levels within a safe range -- despite having access to only current patient data, while doctors used full patient histories. "Our study shows that artificial intelligence can be thoughtfully and responsibly developed to support, rather than replace, the clinical judgment of health care professionals," says co-senior corresponding author Ankit Sakhuja, MBBS, MS, Associate Professor of Medicine (Data-Driven and Digital Medicine) and a member of the Institute for Critical Care Medicine at the Icahn School of Medicine at Mount Sinai. "In complex and high-pressure environments like the ICU, tools like GLUCOSE can provide real-time data-driven guidance tailored to individual patients. This kind of decision support can enhance safety, reduce the risk of complications, and ultimately allow clinicians to focus more of their attention on critical aspects of patient care." The research team trained GLUCOSE using reinforcement learning, which allowed the system to learn optimal decisions through trial and error. They also used advanced methods -- conservative and distributional reinforcement learning -- to ensure the model made cautious, reliable recommendations. The model was then rigorously evaluated and compared to real-world clinical practices. While the results are promising, the researchers caution that GLUCOSE is not intended to replace doctors. It serves as a clinical decision support tool, offering suggestions that physicians can choose to follow based on their judgment and the broader clinical picture. The model could eventually be integrated into electronic health record systems to provide real-time insulin dosing guidance in the ICU, helping reduce complications and improve outcomes. Future steps include adapting the tool for use in other hospital settings, running clinical trials, and exploring ways to integrate it into routine care. One current limitation is that the model does not yet factor in nutrition data, which may affect longer-term glucose control. Still, the ability of GLUCOSE to make accurate recommendations based on limited real-time data highlights its potential to enhance safety and efficiency in postsurgical care. "Our goal is to develop AI systems that meaningfully augment the capabilities of health care providers and ultimately improve patient outcomes," says co-senior corresponding author Girish N. Nadkarni, MD, MPH, Chair of the Windreich Department of Artificial Intelligence and Human Health, Director of the Hasso Plattner Institute for Digital Health, and Irene and Dr. Arthur M. Fishberg Professor of Medicine at the Icahn School of Medicine at Mount Sinai, and Chief AI Officer of the Mount Sinai Health System. "By learning from real-world clinical data and delivering personalized recommendations in real time, models like GLUCOSE represent an important advance toward integrating trustworthy data-driven tools into the clinical workflow. This study offers a glimpse of how AI can be thoughtfully embedded into care to support providers in delivering safer, more precise treatment." The paper is titled "A Distributional Reinforcement Learning Model for Optimal Glucose Control After Cardiac Surgery." The study's authors, as listed in the journal, are Jacob M. Desman, Zhang-Wei Hong, Moein Sabounchi, Ashwin S. Sawant, Jaskirat Gill, Ana C. Costa, Gagan Kumar, Rajeev Sharma, Arpeta Gupta, Paul McCarthy, Veena Nandwani, Doug Powell, Alexandra Carideo, Donnie Goodwin, Sanam Ahmed, Umesh Gidwani, Matthew A. Levin, Robin Varghese, Farzan Filsoufi, Robert Freeman, Avniel Shetreat-Klein, Alexander W. Charney, Ira Hofer, Lili Chan, David Reich, Patricia Kovatch, Roopa Kohli-Seth, Monica Kraft, Pulkit Agrawal, John A. Kellum, Girish N. Nadkarni, and Ankit Sakhuja. The study was funded, in part, by the National Institute of Diabetes and Digestive and Kidney Diseases of the National Institutes of Health grant 5K08DK131286, and by the Clinical and Translational Science Awards (CTSA) grant UL1TR004419 from the National Center for Advancing Translational Sciences. Research reported in this publication was also supported by the Office of Research Infrastructure of the National Institutes of Health under award numbers S10OD026880 and S10OD030463. See the journal paper for conflicts of interest: https://www.nature.com/articles/s41746-025-01709-9. -####- About Mount Sinai's Windreich Department of AI and Human Health Led by Girish N. Nadkarni, MD, MPH -- an international authority on the safe, effective, and ethical use of AI in health care -- Mount Sinai's Windreich Department of AI and Human Health is the first of its kind at a U.S. medical school, pioneering transformative advancements at the intersection of artificial intelligence and human health. The Department is committed to leveraging AI in a responsible, effective, ethical, and safe manner to transform research, clinical care, education, and operations. By bringing together world-class AI expertise, cutting-edge infrastructure, and unparalleled computational power, the department is advancing breakthroughs in multi-scale, multimodal data integration while streamlining pathways for rapid testing and translation into practice. The Department benefits from dynamic collaborations across Mount Sinai, including with the Hasso Plattner Institute for Digital Health at Mount Sinai -- a partnership between the Hasso Plattner Institute for Digital Engineering in Potsdam, Germany, and the Mount Sinai Health System -- which complements its mission by advancing data-driven approaches to improve patient care and health outcomes. At the heart of this innovation is the renowned Icahn School of Medicine at Mount Sinai, which serves as a central hub for learning and collaboration. This unique integration enables dynamic partnerships across institutes, academic departments, hospitals, and outpatient centers, driving progress in disease prevention, improving treatments for complex illnesses, and elevating quality of life on a global scale. In 2024, the Department's innovative NutriScan AI application, developed by the Mount Sinai Health System Clinical Data Science team in partnership with Department faculty, earned Mount Sinai Health System the prestigious Hearst Health Prize. NutriScan is designed to facilitate faster identification and treatment of malnutrition in hospitalized patients. This machine learning tool improves malnutrition diagnosis rates and resource utilization, demonstrating the impactful application of AI in health care. For more information on Mount Sinai's Windreich Department of AI and Human Health, visit: ai.mssm.edu About the Hasso Plattner Institute at Mount Sinai At the Hasso Plattner Institute for Digital Health at Mount Sinai, the tools of data science, biomedical and digital engineering, and medical expertise are used to improve and extend lives. The Institute represents a collaboration between the Hasso Plattner Institute for Digital Engineering in Potsdam, Germany, and the Mount Sinai Health System. Under the leadership of Girish Nadkarni, MD, MPH, who directs the Institute, and Professor Lothar Wieler, a globally recognized expert in public health and digital transformation, they jointly oversee the partnership, driving innovations that positively impact patient lives while transforming how people think about personal health and health systems. The Hasso Plattner Institute for Digital Health at Mount Sinai receives generous support from the Hasso Plattner Foundation. Current research programs and machine learning efforts focus on improving the ability to diagnose and treat patients.