Scoring of myocardial infarction (MI) disease extent in cardiac magnetic resonance (CMR) images has been generally presented in terms of area-based infarct size. However, gradual thinning of the infarcted wall and compensatory hypertrophy of the noninfarcted remote wall during left ventricular (LV) remodeling after MI complicate the accuracy of infarct size measurement. In this study, we measured and compared infarct sizes in mice on late gadolinium enhancement (LGE) images using area-, length-, and radial sector-based methods.MI was induced by permanent ligation of the left coronary artery (n = 6). LGE images were acquired 30 minutes after intravenous injection of Gd-DTPA-BMA. Percentages of infarct size (%Area, %Length, and %Sector) on the LGE images were calculated and compared with histological findings.Infarct sizes obtained by an area-based approach were smaller than those obtained by other measurements. The area-based approach underestimated infarct size compared with the length-based approach. Most infarct sizes measured by each method demonstrated a similar trend, with maximum values determined by sector-based measurements using a mean + SD threshold. Spearman's rank correlation coefficients indicated that the 3 measurements were strongly correlated (P < 0.05) to each other. Significant differences and trends were observed between sector-based infarct sizes with different thresholds when 16 or more sectors were used.In conclusion, our study demonstrated that methods used for the histological calculation of infarct size could be applied to CMR analysis. Moreover, our results showed a similar trend to histological assessment. Sector-based CMR approaches can be useful for infarct size measurement.