Accurately estimating the age of a myocardial infarction (MI) is crucial for determining clinical outcomes and guiding treatment strategies in post-MI patients, such as in arrhythmia risk stratification. This study addresses this challenge by proposing a novel multimodal regression framework for predicting infarct age. The framework integrates key post-MI biomarkers-including myocardial thickness, intramyocardial fat, calcification, and clinical data-as inputs for infarct age estimation. Leveraging computed tomography (CT) imaging, we quantify and analyze the spatial distribution of wall thinning, intramyocardial fat, and calcifications, which are hallmarks of post-MI structural remodeling. These biomarkers, combined with clinical parameters, are fed into a machine learning-based regression model optimized for temporal precision in infarct dating. We leverage multimodal data fusion using late fusion (decision-level) techniques, integrating information from imaging and clinical data through uncertainty quantification. This study evaluates the framework using regression metrics, demonstrating a significant improvement over traditional imputation methods. Additionally, the pathophysiological correlations between intramyocardial fat accumulation, calcification, myocardial remodeling, and clinical factors are explored across various infarct ages, providing valuable insights into the progression of post-MI remodeling. The results suggest that these multimodal inputs offer complementary information that enhances temporal resolution in infarct dating, paving the way for personalized treatment strategies and improved prognostic evaluations in patients with MI.