Masci et al. developed a machine-learning model to estimate cardiovascular biological age (“HeartAge”) from routinely obtainable cardiovascular magnetic resonance (CMR) phenotypes. Using gradient-boosting regression trained on 3,760 healthy UK-Biobank participants, the model was applied to 31,784 individuals to derive the HeartAge-gap, which was defined as the difference between biological and chronological age. HeartAge correlated strongly with chronological age (r = 0.91) and was independent of regression-to-the-mean bias. Across a median follow-up of approximately 5.5 years, each one-year increase in HeartAge-gap was independently associated with a higher risk of a composite cardiovascular outcome in both females and males after adjustment for chronological age and major cardiometabolic confounders. HeartAge-gap also predicted all-cause mortality in females but not in males. External validation in the Multi-Ethnic Study of Atherosclerosis confirmed an association between HeartAge-gap and hard cardiovascular outcomes in females only.