13223
clinical study

Engineering Structural Workflow Efficiencies in the Outpatient Imaging Center: The Synthesis of Human Intervention (HI) and Artificial Intelligence (AI) for Actionable Incidental Findings

Materials & Methods

After IRB approval, AI software was implemented at outpatient imaging centers to flag incidental pulmonary emboli (iPE) and intracranial hemorrhages (ICH). When a case is flagged, a radiology technician notifies the ARNP, who coordinates with the clinical team for confirmation. If validated, the patient is transferred to the emergency department for further care.

Results

Effective training of ARNPs, radiology technicians and coordinators is crucial for optimizing AI-flagged incidental suspected findings. From July 2023 to Feb. 2024, AI significantly reduced turnaround times (TATs) for positive cases: 71.3 minutes for iPEs (vs. 267.6 minutes for negatives, P<0.05) and 69.15 minutes for ICHs (vs. 176.9 minutes for negatives, P<0.05). AI flagged 51 iPEs and 150 ICHs.

Conclusions

This study highlights the potential of point-of-care AI deployment (POC-AID) to enhance incidental suspected finding management in outpatient imaging centers by enabling rapid detection and timely clinical intervention. By integrating AI with human oversight, radiologists can prioritize critical cases, collaborate with healthcare teams and expedite patient referrals, ultimately improving outcomes and healthcare efficiency.

Disclaimer: This links to a third-party website, which may require credentials to review. 

Explore the Latest AI Insights, Trends and Research