Artificial intelligence (AI) in the form of a machine-learned algorithm correctly triaged most postoperative patients to the intensive care unit (ICU) in its first proof-of-concept application in a university hospital setting, according to research presented at Clinical Congress.
The algorithm used in the pilot study included 87 clinical variables and 15 specific criteria related to the appropriateness of admission to the ICU within 48 hours of surgery. An admission to the ICU was considered appropriate if one of the 15 criteria was met. The criteria included: intubation for more than 12 hours, reintubation, respiratory or circulatory arrest, call for rapid response or code, blood pressure below 100/60 mHg for two consecutive hours, heart rate below 60 bpm or above 110 bpm for two consecutive hours, use of pressors, placement of a central venous line or Swan-Ganz catheter, echocardiogram, new onset of cardiac arrhythmia, myocardial infarction, return to the operating room, blood transfusion requiring more than four units, or readmission to the ICU after a prior admission.
The researchers also prepared a questionnaire to prospectively ask clinicians how they would evaluate the need for intensive care for each patient in the study.
“We asked clinicians which is the best pathway for each patient: Should the patient go to the post-acute care unit, a regular floor, or the ICU? We asked the machine the same question and compared the results,” said Francesco Maria Carrano, MD, postdoctoral research fellow, New York University Langone Health, NY, and first author of the study.
AI correctly triaged 41 of the 50 patients in the study (82 percent). Surgeons had an accuracy triage rate of 70 percent (35 patients), intensivists 64 percent (32 patients), and anesthesiologists 58 percent (29 patients). The number of incorrect triage decisions was lowest for AI (18 percent), followed by 30 percent for surgeons, 36 percent for intensivists, and 42 percent for anesthesiologists.
The rate of under-triage was similar for AI (12 percent) and surgeons (10 percent); the rate of over-triage was much lower for AI (6 percent) than for the clinicians, whose rates ranged from 20 percent to 40 percent. Furthermore, AI achieved a positive predictive rate of 50 percent and negative predictive rate of 86 percent.
Clinicians tend to over-triage, meaning if they are in doubt, they err on the side of caution and send a patient to the ICU. However, over-triaging may result in admitting a patient to the ICU who would be better off elsewhere, noted Marcovalerio Melis, MD, FACS, associate professor of surgery, New York University Langone Health, and co-author of the study.
“In those cases, the patient may be unnecessarily exposed to multidrug-resistant bacteria and have an increased overall length of stay,” he said. “On the other hand, under-triaging means a patient who should have been in the ICU is sent to a recovery or step-down unit, and the opportunity for quick rescue of a deteriorating condition is delayed because monitoring is not as intense.”
Although the algorithm in the study clearly outperformed clinicians’ judgment, it is a first step, Dr. Carrano said. The surgical researchers plan to apply the algorithm to other patient populations and include other demographic and clinical features.
“The majority of the patients in this study were men in our hospital. We would like to expand study of the algorithm to women and patients in other hospitals,” Dr. Carrano said.
Dr. Melis noted that the algorithm will be improved and perfected as the machine analyzes more patients, and testing at other sites will validate the AI model.
“Certainly, as shown in this study, the concept is valid and may be extrapolated to any hospital,” he added.