Figure 2

Scatter plots and firs of Ischemic Extent (IE, left-hand y-axis) or Segment-Intensity Response (SIR, right-hand y-axis) vs. Microsphere Flow Ratio (MFR, x-axis), derived from BOLD images acquired at end-systole (A) and end-diastole (B). IE values derived from end-systole BOLD images shows a stronger correlation to an exponential (R²=0.8) than to a linear function (R²=0.7) of MFR, while SIR shows weaker (linear) correlation with MFR (R²=0.5). IE values derived from end-diastole BOLD images shows and equivalent correlation with exponential and linear functions (R²=0.7) of MFR, while SIR shows no correlation to linear or exponential functions of MFR (R²~0). Statistical power at significance level 0.05 was almost 1 for all regressions. These results indicate that the myocardial BOLD effect is more reliably captured with a metric reflecting the size of the affected region (such as the propose IE) than with a metric reflecting mean intensity changes (such as the conventional SIR).