MBIRTorch

MBIR reconstruction implemented in PyTorch.

View on GitHub Full Documentation

Table of Contents

  1. Overview
    1. Key Features
    2. Supported Applications
  2. Installation
    1. Quick install on PyPI via
  3. Quick Start
  4. Related Repositories
  5. Citation
  6. License

Overview

MBIRTorch: Model-Based Iterative Reconstruction (MBIR) for tomographic reconstruction using PyTorch.

Key Features

  • Multiple geometries: parallel beam, cone beam (including curved detector and helical), translation mode, and multi-axis parallel.
  • 4D reconstruction: a time sequence of volumes from a single continuous scan, using multi-agent consensus equilibrium (MACE).
  • Preprocessing routines for NSI and Zeiss scanners, plus geometry calibration.
  • Utilities for metal artifact reduction and stripe removal.
  • Interactive slice and geometry viewers.
  • Informative demos and extensive documentation.
  • Seamless operation on 1 or more GPUs, Mac MPS, or CPU.
  • Compiled torch and Triton kernels for efficiency.

Supported Applications

  • Synchrotron and X-ray CT reconstruction
  • Cone-beam CT (e.g. NorthStar Instrument data)
  • TEM (Transmission Electron Microscopy) reconstruction
  • 2D parallel and fan-beam CT

Installation

Quick install on PyPI via 

pip install mbirtorch

Quick Start

Reconstruct in one line:

import mbirtorch
recon, recon_dict = mbirtorch.recon_simple_parallel(sinogram, angles)
Repo Description
mbirjax_applications CT application demos (NSI cone-beam, view selection)
svmbir Parallel and fan-beam CT reconstruction (CPU-optimized)
mbircone Cone-beam CT reconstruction
OpenMBIR-Index Index of all OpenMBIR packages

Citation

Please use the following BibTeX citation when referencing this software.

@misc{mbirtorch,
  title = {{MBIRTorch}: {H}igh-performance tomographic reconstruction using {PyTorch}},
  author = {Gregery T. Buzzard and Charles A. Bouman and Jingsong Lin and Ziyun Li},
  howpublished = {Software library available from \url{https://github.com/cabouman/mbirtorch}},
  note = {Version 0.1.1},
  year = 2026
}

GitHub’s “Cite this repository” button on the repository page generates this citation from CITATION.cff.

mbirtorch is a PyTorch port of MBIRJAX; please also cite it when referencing the underlying methods.

If you use MBIRJAX in your research, please cite the relevant publications listed in the documentation.

License

BSD 3-Clause License — see LICENSE on GitHub.