Abstract Background Cardiac magnetic resonance (CMR) is currently considered the reference standard for in vivo assessment of myocardial infarction (MI). There is, however, no international consensus on how MI quantification from CMR should be performed. The aim of this study was to test how previously published manual quantification of MI using CMR images compares with MI quantification using a semiautomated, validated method and how this impacts the relationship with MI size estimated by 12-lead electrocardiogram (ECG). Methods Twenty-five patients, from a previously published cohort, were included in the study. All patients had presented with clinical signs of acute coronary syndrome 6 to 12 months before undergoing a CMR examination. The patients had a standard 12-lead ECG recorded at the time of the CMR examination. The previously reported manually quantified MI size was compared with MI size determined using a semiautomated method validated by computer phantom data, experimental in vivo and ex vivo data, and patient data. The MI sizes from the 2 CMR approaches were then compared with the ECG-estimated MI size. Results There was a strong correlation between MI size determined with the 2 CMR methods ( r2 = 0.94, P < .001). There was, however, a systematic overestimation of MI size of approximately 50% by the previously published manually quantified MI size compared with the semiautomated method. This affected the comparison with estimated MI size by ECG, which showed a significant underestimation of MI size compared with manual CMR measurements, but no bias compared with the semiautomated CMR method. Conclusions Manual quantification of MI size by CMR can differ significantly from semiautomated, validated methods taking partial volume effects into account and can lead to erroneous conclusions when compared with ECG.