Objective : To explore an efficient method for the establishment of three-dimensional (3-D) digital models of maxillofacial region including muscle tissue based on CT and MRI images fusion on a personal computer, integration of CT and MRI data, and provide accurate 3D model for biomechanical analysis. Methods : A male volunteer was scanned on maxillofacial region by spiral CT and MRI. Two kinds of data obtained were imported into Mimics 15. In the three sections, namely the transverse, sagittal, coronal sections, two kinds of data were adjusted to the same anatomical layers. The most obvious anatomical points on each layer were selected as registration points. Then, the multi-points registration was implemented for data fusion. Then the bone and facial skin were segmented and 3D reconstructed using CT data, the main facial muscles were segmented and 3D reconstructed using MRI data. Results: The 3D model including 3 pairs of masticatory muscles, 12 pairs of facial expression muscles, facial skin and jaw tissues were established. Conclusion The efficient registration and fusion of CT and MRI datas were accomplished. Moreover, this method can be used for further segmentation and reconstruction of other important structures in craniofacial area, such skin, blood vessel, fat, lymph node and the brain tissues.