New Database of Over 83,000 Surgical Outcomes for Training Artificial Intelligence Is Now Online

Researchers from UCLA and UC Irvine have created a “unique repository of electronic health record data and high-fidelity physiological waveform data from tens of thousands of surgeries that will integrate artificial intelligence to improve patient outcomes,” according to a news article from UCLA Health.

According to the paper describing the database and its uses published in JAMIA Open, the repository has been in the works since 2012 and is intended to fill “a gap in publicly accessible databases that researchers can use to train and test AI algorithms.” It contains data of hospital visits for patients undergoing surgery at UCI Medical Center consisting of “comprehensive electronic health record and high-fidelity physiological waveforms.” Waveforms are data from monitors, like EKGs, that “measure the physiology of the patient either minute by minute or sometimes in real time, for instance during a high-risk surgical procedure.” The data specifically contains general information about each patient’s medical history, including “details about the medical procedure, medicines used, lines or drains utilized during the procedures, and postoperative complications.” It now contains data from around 59,000 patients who underwent 83,500 surgeries. All of the data has been stripped of patient identifiers in accordance with patient privacy laws.

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

Cannesson also states that their “main innovation was to start more than 10 years ago recording these waveforms during surgery.” There is existing precedent for sharing datasets like for patients in the intensive care unit.

The focus in the short-term is to share this information with qualified physicians and researchers. However, a National Institutes of Health initiative called “Bridge to AI” aims to “standardize this data across multiple institutions to eventually create a single repository with the same vocabulary and data architecture.” With AI, Cannesson, the co-lead of the study, says that they expect the data to “develop new algorithms, new predictive tools, to improve the care of surgical patients basically globally.”

UCLA Health has the article.