3D reconstruction of medical images has gained lot of prominence in medical imaging in the recent past. This may be attributed to the differences in the characteristics of the object under consideration and selection of a suitable technique. In this paper reconstruction of brain image from the different views of MRI (Magnetic Resonance Imaging) slices through registration has been discussed. For registering the images Mutual Information (MI) and Cross Correlation (CC) measures have been considered. This work also presents a comparison of the two measures, when used for registration of image. The results indicate that the CC-based method produces better 3D model compared to MI-based method. The Reconstruction was performed separately from axial, sagittal and coronal slices. This work also provides an interface, where from the image reconstructed using a specific view, the other orthogonal slices can be generated using a slicer. The resulting model would be useful for brain image analysis and could also serve as an educational tool for teaching and learning.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

3D Brain MRI Reconstruction: Exploring Axial, Coronal, and Sagittal Views with Interactive Slicing

  • Hema P. Menon,
  • Sanjay S. Menon,
  • Lekshmi S. Nair,
  • Dhanya M. Dhanalakshmy,
  • Gopu Darsan

摘要

3D reconstruction of medical images has gained lot of prominence in medical imaging in the recent past. This may be attributed to the differences in the characteristics of the object under consideration and selection of a suitable technique. In this paper reconstruction of brain image from the different views of MRI (Magnetic Resonance Imaging) slices through registration has been discussed. For registering the images Mutual Information (MI) and Cross Correlation (CC) measures have been considered. This work also presents a comparison of the two measures, when used for registration of image. The results indicate that the CC-based method produces better 3D model compared to MI-based method. The Reconstruction was performed separately from axial, sagittal and coronal slices. This work also provides an interface, where from the image reconstructed using a specific view, the other orthogonal slices can be generated using a slicer. The resulting model would be useful for brain image analysis and could also serve as an educational tool for teaching and learning.