Related Experiment Video
Updated: Jun 26, 2026
11:35
Dual-mode Imaging of Cutaneous Tissue Oxygenation and Vascular Function
Published on: December 8, 2010
Simultaneous registration and intensity normalization of SPECT perfusion images
Margarida Silveira1, Pedro M Q Aguiar
1Institute for Systems and Robotics/IST, Portugal.
Summary
This study introduces a simpler method for processing 3D brain images, simultaneously registering and normalizing SPECT scans. This computationally efficient approach achieves results comparable to complex methods.
Area of Science:
- Medical Imaging
- Neuroimaging
- Computational Biology
Background:
- 3D SPECT brain image analysis requires pre-processing like registration and intensity normalization.
- Current methods often use computationally intensive algorithms, such as Mutual Information (MI)-based registration, before normalization.
Purpose of the Study:
- To propose a computationally simple method for simultaneous registration and intensity normalization of 3D SPECT brain perfusion images.
- To extend a 2D image processing technique to 3D data.
Main Methods:
- A novel approach estimates intensity normalization and registration parameters in alternating steps.
- The method is an extension of a previously proposed 2D image processing technique.
Main Results:
- Experiments with real SPECT images demonstrate the effectiveness of the proposed method.
- The results are comparable to those achieved using more computationally expensive algorithms like MI-based registration.
Conclusions:
- The proposed method offers a computationally efficient alternative for SPECT brain image pre-processing.
- Simultaneous registration and intensity normalization can be achieved effectively with a simpler algorithm.
