Theory of image reconstruction in computed tomography

Area of Science:

• Medical imaging
• Image reconstruction
• Applied mathematics

Background:

• Mathematical methods are fundamental to modern radiographic and radioisotopic imaging.
• Image reconstruction techniques are essential for creating diagnostic images from projection data.

Purpose of the Study:

• To review and classify the primary mathematical methods used in image reconstruction.
• To compare the advantages and limitations of different reconstruction algorithms for specific imaging modalities.

Main Methods:

• Back-projection: Historically significant but largely obsolete due to artifacts.
• Analytical methods: Including Two-dimensional Fourier and Filtered Back-projection, favored for speed and accuracy with ample projection data, especially in X-ray imaging.
• Iterative reconstruction: Preferred for scenarios with limited views, significant noise, or when accounting for factors like gamma-ray attenuation, commonly used in radioisotope imaging.

Main Results:

• Back-projection methods are no longer employed due to significant artifacts.
• Analytical methods offer superior speed and accuracy when numerous projections are available, making them ideal for X-ray imaging.
• Iterative methods demonstrate greater utility and are widely adopted in radioisotope imaging, particularly when dealing with limited data, noise, and physical attenuation effects.

Conclusions:

• The choice of mathematical method for image reconstruction is dictated by the imaging modality and data characteristics.
• Analytical methods are optimal for high-projection-count scenarios like X-ray imaging.
• Iterative methods provide a robust solution for complex situations in radioisotope imaging, addressing limitations of other techniques.