UCLA Releases Public-Access Surgical Outcomes Database for AI Training

Researchers from the University of California, Los Angeles (UCLA) and the University of California, Irvine (UCI) have developed a repository of surgical outcomes data to help the medical research community build new artificial intelligence (AI) algorithms and improve patient outcomes.

The database, known as the Medical Informatics Operating Room Vitals and Events Repository (MOVER), is comprised of electronic health records (EHRs) and high-fidelity physiological waveforms – data from monitors that measure a patient’s physiology either in real time or minute by minute during a medical procedure – from approximately 59,000 patients who underwent 83,500 surgeries.

Specifically, the repository contains information about each patient’s medical and surgical history, including procedure type; medicines, lines, or drains used during surgery; and any postoperative complications. These data may contain insights into patient outcomes that could be useful if incorporated into AI models.

“This information is truly information that physicians and the care team use to make clinical decisions in the acute care setting,” explained Maxime Cannesson, MD, PhD, professor and chair of anesthesiology and perioperative medicine at the David Geffen School of Medicine at UCLA, in the press release. “Before this there was no single repository where a very, very large volume of data that includes the physiological waveforms are accessible to researchers.”

The repository has been in-progress since 2012, and the research team emphasized that the project was undertaken to bridge a gap in perioperative surgery research. The researchers noted that existing publicly available databases that contain surgical outcomes are limited.

“We expect [MOVER] to help the research community to develop new algorithms, new predictive tools, to improve the care of surgical patients basically globally,” Cannesson said. “It’s the first time a surgical database like this has been released. It’s a very wide spectrum of surgeries.”

Moving forward, the research team aims to share the dataset, which is available to researchers who sign a data use agreement (DUA). Further, UCLA is part of the National Institutes of Health (NIH) “Bridge to AI” initiative, under which the MOVER data will be standardized across various institutions in order to create a larger repository.

The research team also underscored that privacy and transparency are key considerations for the repository’s development and growth.

“Patient privacy has been at the forefront of the development of MOVER,” Cannesson noted. “It’s been through a lot of de-identification... There is no patient identifier, no date of surgery. Patients above 90 years old, their age is not available. So it’s been through a lot of de-identification to make sure that no patient identifier is available.”

“The goal is eventually to increase the trust that clinicians and patients have with what you are going to see in the near future – the development of more and more artificial intelligence-based models, especially for the surgical setting,” Cannesson continued.

This research is just one effort to advance AI in perioperative care.

In an August interview with HealthITAnalytics, experts from Mayo Clinic shared how the health system is working to develop AI tools to innovate in the field of organ transplant, particularly in the areas of preventing the need for transplants, improving donor matching, increasing the number of usable organs, preventing organ rejection, and advancing post-transplant care.