| Because the ultrasound imaging has the advantages of no wound, wide applications, operational flexibility, security, reliability, low prices, and so on, it is the most rapidly developed and the quickest popular medical imaging technology after X-ray imaging.There are many clinical application limitations for conventional 2D ultrasound: Physicians need to imagine the 3D organ just from many 2D ultrasound images when diagnosing, it can lead to low efficiency and even errors; 2D ultrasound images only depict a perspective organ, it is difficult to obtain the most optimal imaging plane for the diagnosis; 2D ultrasound images cannot express the surface curvature change of the organ, such as the diagnosis of fetus facial structure, it needs abundant clinical experience of operators and it can be influenced by the subjective factors of operators. 3D ultrasound has many clinical advantages: visualization, it can visually display the 3D structure of the interested region; the accurate spatial orientation, 3D imaging can objectively display the overall structure of organs, then doctors may determine the lesion location accurately and rapidly; the multiple perspectives observation, judge the organ location relations and size changes. Based on the above reasons, the 3D ultrasound imaging technology has been the attention focus of clinical user and engineering academia.After development of nearly 20 years, there are four types of 3D US system on the digitized medicine phantom equipment market: mechanical arm 3D US system, sensor localization freehand 3D US system, non-sensor localization freehand 3D US system, 2D Paneled probe freehand 3D US system. Comparatively speaking, the sensor localization freehand 3D US system has the prominent advantages: using the conventional 1D linear array probe (low cost), accurate spatial information of images (This is the basic prerequisite of realizing image survey or quantitative analysis technology).Freehand 3D US imaging system has an important step, is 3D volume data reconstruction, which may also be called as 3D volume data interpolation. It obtains some 2D images through the ultrasound diagnosis system from several different positions of human body parts(organs), then inputs these 2D images and the position, angle information into the computer, they are compounded and processed by computer (including the interpolation-smoothing for the spacing of the adjacent plane, forming a rule 3D volume data), the process of forming rule 3D volume data from 2D images through computer processing is the 3D volume data Reconstruction. The calculation amount of Reconstruction Algorithm is very large, therefore, the key of volume data reconstruction is: improving the reconstruction speed under the premise of guaranteeing the reconstruction quality. This paper has realized the parallel processing of volume data reconstruction based on OpenMP under the multi-core CPU environment, it greatly raises the reconstruction speed and is advantageous for real-time visualization of freehand 3D imaging system.RBF has widespread applications in the field of scattered data interpolation, and has been research hotspot, because it can guarantee that the corresponding linear coefficient matrix equations are reversible even if the data points are very disorderly, the reconstruction effect is also better. The motion of probe is controlled by the physician in the freehand ultrasound imaging, therefore, the relative distance and direction of the 2D US image is arbitrary. This means that the pixels are arranged irregularly in the voxel array. Therefore, interpolation problem can be classified as unstructured or scattered data interpolation. To further improve the reconstruction quality, favor the visualization analysis, raise the diagnosis level, this paper in-depth, or comprehensive studied the RBF principle and has realized RBF interpolation algorithm, made the better reconstruction effect. |
PDF Full Text Request