Overview Despite consensus on a clear set of effective kidney-protective strategies to reduce the risk of acute kidney injury (AKI), these treatments are often underutilized during surgery because perioperative risk of AKI is underestimated; as a result, AKI affects 18-47% of surgical patients and is associated with adverse outcomes including chronic kidney disease, major adverse cardiovascular events (MACE), increased healthcare costs, and death. This problem provides an excellent opportunity for machine learning (ML) because ML has already been shown to predict AKI better than clinicians, but these models have not yet been implemented in the perioperative period because of multiple technical and logistical barriers that we have developed technology to overcome. We are implementing and updating innovative models that predict and alert clinicians of impending perioperative AKI during surgery when preventative treatments can be provided and, in the process, use these models to better understand the landscape of risk and design new treatments for AKI. Affiliated Lab AI Clinical Innovation Lab Principal investigator UCSF Andrew Bishara, MD Assistant Professor External persons Ӧzlem S. Çakmakkaya MD PhD Mentors UCSF Romain Pirracchio, MD, PhD Professor, Chief, ZSFG Anesthesia and Perioperative Care External persons Atul Butte, MD, PhD Professor, Pediatrics Kathleen Liu Collaborators External persons Allan Basbaum, PhD Chair and Professor Edward Hsiao, MD, PhD Professor Seeking collaborators Our project is looking for individuals, specifically those who would like to study AI in the healthcare setting. Those interested in working on this project should have a technical background. Email Us Support this research Are you excited by the innovative work we’re doing on this project? Learn how your financial support can make the difference in our work. Support
