AbstractObjectiveTo identify, characterize, and automatically classify hypoperfusion-related changes in the blood oxygenation level dependent (BOLD) signal in acute stroke using spatial independent component analysis of resting-state functional MRI data.MethodsWe applied spatial independent component analysis to resting-state functional MRI data of 37 stroke patients scanned within 24 hours of symptom onset, 17 of whom received follow-up scans the next day. All patients also received dynamic susceptibility contrast MRI. After denoising and manually classifying the components, we extracted a set of temporal and spatial features from each independent component and used a generalized linear model to automatically identify components related to tissue hypoperfusion.ResultsOur analysis revealed “Hypoperfusion spatially-Independent Components” (HICs) whose BOLD signal spatial patterns resembled regions of delayed perfusion depicted by dynamic susceptibility contrast MRI. These HICs were detected even in the presence of excessive patient motion, and disappeared following successful tissue reperfusion. The unique spatial and temporal features of HICs allowed them to be distinguished with high accuracy from other components in a user-independent manner (AUC = 0.95, accuracy = 0.96, sensitivity = 1.00, specificity = 0.96).InterpretationOur study presents a new, non-invasive method for assessing blood flow in acute stroke that minimizes interpretative subjectivity and is robust to severe patient motion.
Vascular change measured with independent component analysis of dynamic susceptibility contrast MRI predicts bevacizumab response in high-grade glioma
