Clot-based radiomics features predict first pass effect in acute ischemic stroke.

Area of Science:

• Radiology
• Medical Imaging
• Stroke Neurology

Background:

• Acute ischemic stroke (AIS) requires rapid intervention, often mechanical thrombectomy (MT).
• Predicting the first pass effect (FPE), a measure of successful reperfusion, is crucial for treatment planning.
• Clot characteristics on imaging may influence MT outcomes.

Purpose of the Study:

• To evaluate clot-based radiomics features (RFs) for predicting FPE in AIS patients.
• To identify other clinical and imaging variables associated with FPE.

Main Methods:

• Retrospective review of 52 AIS patients undergoing MT for anterior circulation large vessel occlusion.
• Extraction of RFs from pre-treatment noncontrast computed tomography (NCCT) clot images.
• Logistic regression analysis to identify independent predictors of FPE.

Main Results:

• 48.1% of patients achieved FPE.
• Twelve RFs differed significantly between FPE and non-FPE groups.
• Long-run low gray-level emphasis and zone percentage were independent predictors of FPE, alongside female sex and baseline ASPECT score >8.5.
• A predictive model using these factors achieved 83% diagnostic accuracy for FPE.

Conclusions:

• Clot-based radiomics features on NCCT show promise in estimating MT success (FPE) in AIS patients.
• Integrating RFs with clinical factors enhances FPE prediction.