3D Medical Image Visualization Platform Based On Volume Rendering

3D medical image visualization is an important branch of volume visualization. It rapidly developed with computer tomography, NMR imaging, ultrasound and other medical imaging technology. 3D medical image visualization obtains the two-dimensional digital image sequence of human tissues, and reconstructs it to volume data. volume data provides more complete information for medical diagnosis, surgical planning and treatment monitoring. Because the 3D volume visualization technology can provide visual expression of volume data, it has become an important research field of medical image visualization. How to represent the information in volume data accurately and intuitively is one of the main tasks of 3D medical image visualization.VolPack is an experimental library developed on the basis of Shear-Warp Factorization. But this study found that it lacks of clarity in module definition, especially in algorithm implementation it is difficult to read and understand. Therefore, based on the thinking of refactor, this study achieved a C++ application framework with Factory Method pattern, we named it VolPlus. VolPlus framework inherits many advantages of the object-oriented program design. It is well structured and can be easily expanded.Then based on the layering design of software engineering, this study built an MFC application on VolPlus framework. It is an essential 3D medical image Visualization platform. The platform has proved the advantages of easy expansion of VolPlus framework, and the platform is also has a good user interface. Dragging the mouse in the screen, we can instantly observe the human tissues from any direction. In computer-assisted surgery and image-guided treatment, the 3D medical image visualization platform can be applied.Next, based on the human anatomy assumption, this study found that the voxels in volume data are a direct reflection of human tissue density distribution. It is possible to show or hide certain human tissue according to human tissues density histogram. Therefore, we can segment different human tissue by modifying volume data directly. It is a simple, intuitive, rapid method for virtual segmentation in 3D medical image visualization.Finally, based on the research of virtual segmentation, we can also hide part of the human tissue on the specified location and direction. This study implemented virtual anatomy by giving zero value to these voxels we don't need. After such process, we can get the three-dimensional medical images with anatomy effects. It is very useful in medical diagnosis.From the researches above, this study found that the density information of volume data of human tissues is an important basis for virtual segmentation. In addition, process of volume data is an effective method for virtual anatomy. So we can preprocess the volume data before render it on the 3D medical image visualization platform, then we can more in-depth observe the anatomy of human tissues structure. This study further expands application of Volume Visualization in medical diagnosis, surgical planning and treatment monitoring.