AI assisted reader evaluation in acute CT head interpretation (AI-REACT): protocol for a multireader multicase study
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.This study evaluates an AI tool's impact on healthcare staff interpreting non-contrast CT head scans (NCCTH). It assesses accuracy and review time changes when AI assists clinicians and radiographers in detecting abnormalities.
Area Of Science
- Radiology and Medical Imaging
- Artificial Intelligence in Healthcare
- Clinical Decision Support Systems
Background
- Non-contrast CT head scans (NCCTH) are crucial in emergency departments.
- Artificial intelligence (AI) tools are emerging for NCCTH abnormality detection.
- Limited evidence exists on AI's impact on non-radiologist healthcare staff reviewing NCCTH.
Purpose Of The Study
- To assess the standalone performance of the qER EU 2.0 AI tool on NCCTH.
- To evaluate the impact of AI assistance on the diagnostic accuracy and review time of various healthcare professionals interpreting NCCTH.
- To analyze diagnostic confidence changes with AI tool utilization.
Main Methods
- Retrospective analysis of 150 NCCTH (60 control, 90 with abnormalities).
- 30 healthcare professionals (radiologists, emergency clinicians, CT radiographers) reviewed scans with and without AI assistance.
- Performance measured by accuracy (AUC), review time, and diagnostic confidence, compared against neuroradiologist ground truth.
Main Results
- The study will assess the standalone performance of the qER EU 2.0 AI tool.
- It will quantify the change in accuracy (AUC), median review time, and self-reported diagnostic confidence for readers using the AI tool.
- Subgroup analyses will explore variations by professional group, seniority, and pathology.
Conclusions
- The findings will provide evidence on the effectiveness of AI tools in supporting clinical decision-making for NCCTH interpretation.
- This research will inform the integration of AI into routine clinical practice for emergency imaging.
- The study aims to clarify the real-world impact of AI on diverse healthcare staff reviewing NCCTH.
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
