2-022 Artifact-corrected artificial intelligence-LED intracoronary oct analysis identifies plaque progression and vulnerability, drug efficacy and patient events
Artifact-corrected AI analyzes intracoronary OCT data to identify plaque progression and cardiovascular event risks, suggesting improved clinical utility.
Key Points
AutoOCT achieved an 83% plaque classification accuracy, comparable to expert analysis, facilitating timely patient management.
Pre-processing techniques improved artifact correction by over 70%, enhancing the detection of fibrous plaques in OCT images.
Analysis revealed AutoOCT effectively identified high-risk plaque features predictive of major adverse cardiovascular events.
The system's capabilities indicate potential for real-time support in clinical trials assessing drug efficacy and plaque stability.