Cardiac MRI data from 1,789 adults across three European population cohorts (SHIP-TREND-0, SHIP-START-2, KORA-FF4) reveal that regional left ventricular wall thickness patterns — mapped via k-means clustering — identify distinct cardiovascular risk profiles invisible to conventional remodeling metrics. In women, the highest-risk cluster correlated with a 10.6 percentage-point increase in 10-year Framingham Risk Score, independent of blood pressure, myocardial mass, average wall thickness, and LV concentricity — a finding not replicated with the same independence in men.
Standard LV remodeling classifications rely on blunt categorizations — concentric vs. eccentric hypertrophy — that flatten the spatial complexity of myocardial adaptation. This work advances a spatially granular approach that could meaningfully close the diagnostic gap for women, who are historically underdiagnosed for structural heart disease partly because their remodeling patterns differ from male-derived reference norms. The sex-specific clustering findings are particularly notable given that women in this cohort had far lower baseline CVD prevalence (2% vs. 7.8%), suggesting the regional patterning captures subclinical risk earlier.
Limitations are significant: cohorts skew toward middle-aged European adults, limiting generalizability; k-means clustering is sensitive to initialization and may not generalize across imaging protocols; and causal inference is constrained by observational design. Critically, this is a preprint posted on medRxiv and has not yet been peer-reviewed — findings should be considered preliminary. If validated, automated regional wall analysis integrated into clinical CMR workflows could represent a meaningful upgrade to women's cardiovascular risk stratification.
