Motivated by the growing wealth of biological data and the increasing pathophysiological knowledge at different scales, spectacular progress have been made in comprehensive modeling to understand the complex nature of biological organs, such as the heart. Today, the complexity of cardiac electromechanical models is outstanding; they couple multiple scales, from a genetic level to cellular, organ, and until including whole body interactions, and they already proved their importance to understand the pathophysiology of human diseases such as heart failure and dyssynchrony (Kerckhoffs et al., 2008). However these models fail to faithfully reproduce the behavior of an individual heart. This last decade, a relatively novel approach has emerged as an excellent answer to that drawback: the specific modeling! With the continual grow of computing power; we strongly believe that we now have reached a state where specific models are increasingly feasible and almost ready to be used clinically to assist the physicians. In this study, we present the methodology developed to design a realistic whole heart dog-specific model that includes the four most important components for models of cardiac electromechanics: anatomical, electrophysiology, mechanics and hemodynamics. To specifically fit these components into the model, mechanical anisotropy of the myocardium, active and passive tissue properties, and muscle activation sequence were measured during canine experimental studies. Additionally, high-end imaging tools such as Magnetic Resonance Imaging, Diffusion Tensor Imaging, and micro Computed Tomography were performed on each animal to acquire specific morphological details on anatomy and muscle fibers orientation. Collectively, the methods presented in this paper will lay the framework toward the development of cardiac patient-specific models. These models will be invaluable clinically to predict and optimize the outcome of therapies and interventional surgery on patients