Three-dimensional reconstruction and fusion for multi-modality spinal images

Yuan Tsung Chen, Ming Shi Wang

研究成果: Article同行評審

22 引文 斯高帕斯(Scopus)


Medical diagnosis can benefit from the complementary information in different modality images. Multi-modal image registration and fusion is an essential task in numerous three-dimensional (3D) medical image-processing applications. Registered images are not only providing more correlative information to aid in diagnosis, but also assisting with the planning and monitoring of both surgery and radiotherapy. This research is directed at registering different images captured from Computed Tomography (CT) and Magnetic Resonance (MR) imaging devices, respectively, to acquire more thorough information for disease diagnosis. Because MR bone model segmentation is difficult, this research used a 3D model obtained from CT images. This model accomplishes image registration by optimizing the gradient information accumulated around the bony boundary areas with respect to the 3D model. This system involves pre-processing, 2D segmentation, 3D registration, fusion and sub-system rendering. This method provides desired image operation, robustness verification, and multi-modality spinal image registration accuracy. The proposed system is useful in observing the foramen and nerve root. Because the registration can be performed without external markers, a better choice for clinical usage is provided for lumbar spine diagnosis.

頁(從 - 到)21-31
期刊Computerized Medical Imaging and Graphics
出版狀態Published - 2004

All Science Journal Classification (ASJC) codes

  • 放射與超音波技術
  • 放射學、核子醫學和影像學
  • 電腦視覺和模式識別
  • 健康資訊學
  • 電腦繪圖與電腦輔助設計


深入研究「Three-dimensional reconstruction and fusion for multi-modality spinal images」主題。共同形成了獨特的指紋。