Gray-Matter Volume Estimate Score: A Novel Semi-Automatic Method Measuring Early Ischemic Change on CT

Area of Science:

• Neuroradiology
• Stroke Imaging
• Quantitative Imaging Biomarkers

Background:

• Early ischemic changes on CT are crucial for stroke management.
• Existing scoring systems like ASPECTS have limitations in accuracy and consistency.
• A novel semi-automated method, GRAVES, was developed to quantify early ischemic changes.

Purpose of the Study:

• To compare the novel GRAVES score with the established ASPECTS for outcome prediction in ischemic stroke.
• To evaluate the inter-rater agreement between GRAVES and ASPECTS.
• To assess the utility of GRAVES in predicting outcomes after intra-arterial therapy (IAT).

Main Methods:

• Retrospective cohort study of 94 patients with anterior circulation ischemic stroke treated with IAT.
• Independent assessment of pretreatment CT scans using both GRAVES and ASPECTS by two neurologists.
• GRAVES defined as estimated hypodense lesion percentage in gray matter.
• Statistical analysis included Spearman correlation, ROC comparison, and ICC comparison.

Main Results:

• GRAVES showed good correlation with ASPECTS (Spearman's rho = 0.642, P<0.001).
• GRAVES demonstrated comparable predictive value for favorable outcomes to ASPECTS (AUC 0.765 vs. 0.717, P=0.308).
• GRAVES exhibited significantly better inter-rater agreement than ASPECTS (ICC 0.978 vs. 0.895, P<0.001).

Conclusions:

• GRAVES is a reliable method for semi-automatically quantifying early ischemic changes on CT.
• GRAVES performs comparably to ASPECTS in predicting clinical outcomes after IAT.
• GRAVES offers improved inter-rater reliability, making it a potentially valuable tool in stroke assessment.