AI-based exam queue tool prioritizes CT scans in the ED
New research suggests that artificial intelligence-based CT scheduling systems could significantly reduce imaging delays for emergency department patients with concerning symptoms.
A new paper in Radiology: Artificial Intelligence details how an AI-driven CT queue prioritization system can improve patients’ time to diagnosis without requiring additional imaging equipment or staffing. The simulation analysis found that use of the system could reduce wait times by nearly an hour for patients most likely to have critical findings.
The retrospective study analyzed data from nearly 314,000 emergency department visits between 2020 and 2024 to develop a machine learning model that predicts at the time the scan is ordered whether it is likely reveal clinically actionable findings. Researchers then used a discrete-event simulation calibrated to real-world emergency department operations to compare the AI-driven prioritization strategy with the conventional first-in, first-out approach across an additional nearly 21,000 CT exams.
Compared with the standard workflow, the AI queue reduced median wait times for actionable CT studies by 10.75 minutes and decreased 90th-percentile (most actionable) wait times by 43.36 minutes. Actionable examinations completed within one hour increased from 48% to 57%. Although the model modestly increased 90th-percentile wait times for nonactionable studies by 14.46 minutes, median wait times for those patients also decreased by 5.8 minutes. The model achieved 80% to 87% of the maximum performance possible with perfect predictions.
Corresponding author David Kim, MD, PhD, with the department of emergency medicine at Stanford University, and co-authors suggested their findings highlight the potential of similar tools to improve both patient care and clinical workflows without requiring a significant hardware installation.
“AI-driven CT imaging queue prioritization can reduce the time to diagnosis for critical findings with minimal disruption to lower-acuity patients, using existing data streams and without the need for additional hardware,” the team concluded.
Read more here.
In addition to her background in journalism, Hannah also has patient-facing experience in clinical settings, having spent more than 12 years working as a registered rad tech. She began covering the medical imaging industry for Innovate Healthcare in 2021.