For better management of cardiovascular disease, it is of utmost importance to categorize the subjects into different risk groups. This categorization can be made based on cardiovascular risk factors including the family history of the subject. Imaging techniques play an increasing role in order to assess ardiovascular risk factors. In this thesis we set out to develop and evaluate automatic techniques for the extraction of quantitative imaging biomarkers for coronary artery disease (CAD). One of the important cardiovascular risk factor is the presence of calcium in the arteries. We presented an automatic method that can compute the amount of calcium scores for the whole heart as well as for each of the coronary arteries from CT data. The system also categorizes patients into different risk groups. This vessel specific calcium lesion information can be used for treatment planning and assessing progression of CAD in follow up studies. The possibility to assign calcium to individual coronary arteries was possible owing to the ’Coronary Density Estimate’. The second imaging biomarker is epicardial fat volume. We resent a method that can accurately quantify the amount of epicardial fat volume. It was demonstrated that the method performs as good as the manual observers, hence has great potential to be used in daily clinical practice. In a clinical study on 2298 subjects it was demonstrated that indeed larger volumes of epicardial fat volumes were related to larger volumes of calcified lesions in the various vessel beds. The potential of this biomarker will need to be established in multiple larger studies. The third imaging biomarker in CAD considered in this thesis is coronary artery stenosis grade. Accurate detection and quantification of coronary stenoses is of great importance, as this information is very important for the clinician in order to make accurate treatment selection and planning. We investigated the ability of detecting and quantifying coronary stenoses from CTA data. We demonstrated that the vessel lumen can be segmented with a precision similar to the human observers, but that it is still a challenge to be able to distinguish between significant and non-significant lesions. Quantitative imaging biomarkers in CAD may provide both anatomical and functional information, and are often obtained from different imaging modalities. An important subject with respect to treatment planning is therefore the ability to combine information from different modalities in an integrated display. The SMARTVis system was introduced to fuse anatomical information from CTA scans and functional information from SPECT-MPI into one display. The integrated visualization proposed in the SMARTVis system enables a one-stop-shop visual exploration of cardiac anatomical and functional data, to maximally exploit the complementary information of multiple imaging modalities. It has been confirmed that such comprehensive visualizations allow to effectively relate perfusion defects and coronary lesions, and that fused integrated analysis leads to a more accurate diagnosis. Automatic image processing plays an increasingly important role. Not only to extract relevant quantitative imaging biomarkers from CT imaging data, but also establish with what accuracy they can be assessed. For a number of relevant cardiovascular quantitative imaging biomarkers, this thesis has provided the required methodology.