差分累积PET/CT三维融合

PET/CT一体机的出现,使PET/CT三维融合方法成为研究的热点。现有基于光线投射算法的PET/CT三维融合需要构建多个转换函数,人机交互复杂且耗时较多,融合效果中正常与非正常组织对比度差。针对这一问题,提出了一种基于差分累积的PET/CT三维融合方法,沿视线方向,根据PET数据中采样点灰度值的变化,计算其对成像平面的贡献和融合权重;对CT数据进行光照计算后,根据计算得到的权重将PET与CT当前采样点的颜色和不透明度进行融合。实验结果表明,该方法的人机交互简便,PET/CT三维融合后的渲染结果中可疑病灶组织与正常组织有较好的对比度。

Dual-Modality PET/CT Visualization Using Difference Accumulation

With the emergence of the dual-modality medical imaging scanners such as PET/CT, three-dimensional visualization of PET/CT data is becoming a research focus. Existing three-dimensional fusion based ray-casting algorithm need to build more than one transfer function, human-computer interaction is complex and time-consuming, the rendering result have low contrast. To address the problems, we propose a new three-dimensional PET/CT data fusion algorithm based difference accumulation. For the present method, along the ray direction, according to the intensity change of the sampling points in the PET data, calculate the contribution of the imaging plane and the fusion weight. For the CT data, implement illumination calculation, then fuse the PET and CT data based on the obtained weight. The experimental results show that the present method could achieve high contrast PET-CT data fusion results with simple human-computer interaction.