Tumor motion during irradiation reduces target coverage and increases dose to healthy tissues. Prediction of respiratory motion has the potential to substantially improve cancer radiation therapy. The respiratory motion is complex and its prediction is not a simple task, especially that breathing is controlled by the independent action of the diaphragm muscles and thorax. The diaphragm is the principal muscle used in the process of respiration and its modeling is essential for assessing the respiratory motion. In this context, an accurate patient-specific finite element(FE) based biomechanical model can be used to predict diaphragm deformation. In this paper, we have developed a FE model of the respiratory system including the diaphragm behavior and the complete thorax with musculoskeletal structure. These incorporate the ribs kinematics extracted directly from the Computed Tomography (CT) scan images. In order to demonstrate the effectiveness of our biomechanical model, a qualitative and quantitative comparison between the FE simulations and the CT scan images were performed. Upon application of linear elastic models, our results show that a linear elastic model can accurately predict diaphragm deformations. These comparisons demonstrate the effectiveness of the proposed physically-based model. The developed computational model could be a valuable tool for respiratory system deformation prediction in order to be controlled and monitored by external sensors during the treatment .