Off-console automated artificial intelligence enhanced workflow enables improved emergency department CT capacity.

Area of Science:

• Radiology and Medical Imaging
• Health Informatics
• Artificial Intelligence in Healthcare

Background:

• Rising emergency department (ED) demand necessitates increased computed tomography (CT) capacity.
• Conventional CT workflows present inherent inefficiencies and limitations.
• Optimizing CT throughput is crucial for timely patient care in emergency settings.

Purpose of the Study:

• To evaluate an automated, off-console AI-enhanced workflow for CT image processing.
• To determine if the off-console workflow reduces overall ED CT exam begin-to-end times.
• To assess the impact of the off-console workflow on CT scanner occupancy time and its variability.

Main Methods:

• A retrospective study comparing conventional and off-console CT workflows over 3-month periods.
• Inclusion of all adult ED patients undergoing CT exams.
• Collection of exam begin and end timestamps from the Electronic Medical Record (EMR).
• Analysis of four specific subgroups of Emergency Severity Index (ESI) level 1 patients undergoing common CT exams.

Main Results:

• The off-console workflow reduced median exam begin-to-end times by 36% (P < 0.001).
• Significant decreases in median CT occupied time (56-75%) were observed in four subgroups (P < 0.001).
• Reduced variability in CT occupied time was noted in three of the four subgroups.

Conclusions:

• The implemented off-console workflow effectively increases CT capacity to meet ED demand.
• Broad implementation of this AI-enhanced workflow could yield similar improvements across various imaging settings and exam types.
• This approach offers a scalable solution for enhancing efficiency in diagnostic imaging departments.