Tomosipo: fast, flexible, and convenient 3D tomography for complex scanning geometries in Python

Area of Science:

• Medical imaging
• Computational imaging
• Image reconstruction

Background:

• Tomography reconstructs object interiors from projection images.
• Standard acquisition geometries (circular, helical) are common.
• Complex geometries require specialized software.

Purpose of the Study:

• To introduce tomosipo, a Python package for complex tomography.
• To provide concise representation, visualization, and computation for advanced geometries.
• To address limitations of current software for non-standard acquisition paths.

Main Methods:

• Development of the tomosipo Python package.
• Implementation of tools for representing complex acquisition geometries.
• Demonstration using case studies.

Main Results:

• tomosipo offers a flexible and intuitive solution.
• The package handles complex geometries effectively.
• Case studies validate its power and flexibility.

Conclusions:

• tomosipo simplifies reconstruction for complex tomography.
• The software enhances flexibility in advanced imaging.
• It is a valuable tool for researchers and practitioners.