Previous chapters deal with analytic image reconstruction algorithms. This chapter, on the other hand, introduces iterative image reconstruction algorithms. Due to high speed computers, iterative algorithms get more and more attention in medical image reconstruction. This chapter describes the imaging problem as a system of linear equations, and reconstructs an image by minimizing an objective function. Many algorithms are available to solve the system of linear equations or to minimize an objective function. The objective function can be set up by using the likelihood function, and can also include the prior knowledge about the image. The likelihood function models the noise distribution in the projection measurements. The ML-EM algorithm or OS-EM algorithm is the most popular iterative image reconstruction algorithm in emission tomography, and this chapter has devoted significant efforts to it. Many strategies for noise control are discussed. This chapter also presents a recent research hot spot—image reconstruction with highly undersampled data, which is often referred to as compressed sensing and is, in fact, nothing but another application of Bayesian image reconstruction.
Weitere Kapitel dieses Buchs durch Wischen aufrufen
Bitte loggen Sie sich ein, um Zugang zu diesem Inhalt zu erhalten
Sie möchten Zugang zu diesem Inhalt erhalten? Dann informieren Sie sich jetzt über unsere Produkte:
- Iterative Reconstruction
Prof. Dr. Gengsheng Lawrence Zeng
- Springer Berlin Heidelberg
Neuer Inhalt/© Filograph | Getty Images | iStock