Using Discrete Event Simulation to Design and Assess an AI-aided Workflow for Same-day Diagnostic Testing of Women...

Area of Science:

• Radiology
• Medical Imaging
• Artificial Intelligence in Healthcare

Background:

• Screening mammograms can lead to patient anxiety due to lengthy diagnostic workup processes.
• Artificial intelligence (AI) interpretation of mammograms shows potential to decrease unnecessary patient recalls.

Purpose of the Study:

• To evaluate the unintended consequences of an AI-aided interpretation and same-day diagnostic workflow in a high-volume breast imaging center.
• To assess the impact of this new workflow on patient volume, patient time in clinic, and waiting times.

Main Methods:

• Discrete event simulation was employed to model the proposed AI-aided interpretation and same-day diagnostic workup.
• The simulation analyzed the effects on key operational metrics within a breast imaging setting.

Main Results:

• The AI-aided workflow is projected to reduce daily patient volume by 4%.
• Patient time spent at the clinic is expected to increase by 24%.
• Waiting times are estimated to increase by 13-31%.

Conclusions:

• While AI in mammography can streamline some aspects, the proposed same-day diagnostic workflow may introduce operational challenges.
• Strategies such as adjusting operating hours and increasing resources (equipment, personnel) are suggested to mitigate negative impacts.