High Performance Computing
The goal of this project is to develop implementations of critical applications and to develop novel algorithms to solve key problems across a broad spectrum of medical image analysis areas, including registration, segmentation, and visualization. When necessary, portable and efficient implementations will be developed using standard parallel software technologies so as to be able to exploit the computational power of symmetric multiprocessor (SMP) machines and clusters. This approach will enable the rapid analysis of enormous data sets while maintaining portability and flexibility.
One of the unique characteristics of image-guided therapy is that the time value of a computation is extremely limited. A focus of our research efforts is the development of parallel algorithms that are executed during image-guided therapy and for which the result must be achieved at a rate compatible with surgical decision making. A major objective is the creation of parallel algorithms and fast implementations on parallel hardware.
Back to Research Projects.
- MRT Data Dissemination
- MRT Data Mining
- Improved targeting in minimally invasive image-guided radio frequency ablation
- Quantitative assessment of RFA procedures outcome
- MRI/PET/CT guided biopsy
- Compensation for geometrical distortion on intra-operative MRI
- Quantification of WMT displacement due to brain shift
- Development of novel framework for brain tumor segmentation on MRI