PySimPace
| Developer(s) | Snehil Kumar |
|---|---|
| Initial release | 2025 |
| Stable release | 2.0.1
/ 10 June 2025 |
| Repository | https://github.com/snehil03july/py-SimPace |
| Written in | Python |
| Engine | |
| Operating system | Cross-platform |
| Platform | Python |
| Type | Medical imaging software, scientific software, neuroimaging software |
| License | MIT License |
| Website | https://pypi.org/project/py-simpace/ |
Search PySimPace on Amazon.
PySimPace is an open-source Python software package for simulating realistic motion artifacts and image artifacts in magnetic resonance imaging (MRI) data. The toolkit supports both structural MRI (3D) and functional MRI (fMRI, 4D), and is designed for medical imaging research, machine learning dataset generation, and the development and evaluation of MRI motion correction algorithms.[1][2]
The software was presented as PySimPace v2.0: An Easy-to-Use Simulation Tool with Machine Learning Pipelines for Realistic MRI Motion Artifact Generation at the 33rd ACM International Conference on Multimedia in 2025.[3]
Overview
Motion artifacts are a common problem in MRI and fMRI because patient movement during scanning can degrade image quality and reduce the reliability of downstream clinical or research analysis. PySimPace provides a programmable framework for generating motion-corrupted MRI data from clean scans. It is intended to help researchers create paired clean and corrupted datasets for supervised learning, benchmark motion correction methods, and study the effect of controlled artifact generation on medical imaging pipelines.[4]
The package includes methods for image-space and k-space artifact simulation, ghosting, Gibbs ringing, physiological noise generation, and machine-learning-oriented paired dataset creation.[1]
History
PySimPace was developed at the University of Exeter by Snehil Kumar, Neil Vaughan, Zeyu Fu, and Heather Wilson. The software was created to support research into retrospective MRI and fMRI motion correction, particularly in the context of machine learning and generative artificial intelligence methods for medical imaging.[3]
Version 2.0 of PySimPace was accepted in the Open Source Software track of ACM Multimedia 2025.[5] The corresponding conference paper describes PySimPace as a pip-installable toolkit integrating MRI motion simulation with machine learning pipelines.[3]
Features
PySimPace includes support for:
- structural MRI artifact simulation;
- functional MRI artifact simulation;
- image-space motion simulation;
- k-space motion simulation;
- blended k-space motion simulation;
- ghosting artifact generation;
- Gibbs ringing simulation;
- physiological noise generation;
- paired clean and corrupted dataset generation;
- machine learning pipeline integration;
- NIfTI image input and output;
- command-line interface tools;
- tutorial notebooks and example workflows.[1][2]
Design and architecture
PySimPace is structured as a modular Python toolkit. The package separates structural MRI simulation, fMRI simulation, transformations, noise modelling, input/output handling, analysis, and machine learning utilities into distinct software components.[2]
The software is designed to allow researchers to define motion parameters, apply controlled corruption to MRI volumes, and generate datasets suitable for algorithm development. This modular design allows the package to be used both as a standalone simulation tool and as part of larger machine learning workflows.[4]
Applications
PySimPace is intended for use in:
- medical image analysis research;
- MRI and fMRI motion artifact simulation;
- development of motion correction algorithms;
- training deep learning models for artifact removal;
- data augmentation for neuroimaging;
- benchmarking MRI reconstruction and correction methods;
- teaching and reproducible experimentation in medical imaging.
Because the software can generate paired clean and corrupted MRI data, it can be used in supervised learning settings where a model requires both an artifact-free reference and a motion-corrupted input.[1]
Installation
PySimPace is distributed through the Python Package Index. It can be installed using pip:
pip install py-simpace
The source code is available on GitHub under the MIT License.[2]
Publication
The PySimPace v2.0 paper was published in the proceedings of ACM Multimedia 2025. The paper is also available through Open Research Exeter, the University of Exeter's institutional research repository.[3][4]
See also
- Magnetic resonance imaging
- Functional magnetic resonance imaging
- Medical image computing
- Neuroimaging
- Scientific Python
- Machine learning in bioinformatics
- Medical image analysis
References
- ↑ 1.0 1.1 1.2 1.3 "py-simpace". Python Package Index. Retrieved 24 June 2026.
- ↑ 2.0 2.1 2.2 2.3 "snehil03july/py-SimPace". GitHub. Retrieved 24 June 2026.
- ↑ 3.0 3.1 3.2 3.3 Kumar, Snehil; Vaughan, Neil; Fu, Zeyu; Wilson, Heather (2025). PySimPace v2.0: An Easy-to-Use Simulation Tool with Machine Learning Pipelines for Realistic MRI Motion Artifact Generation. MM '25: Proceedings of the 33rd ACM International Conference on Multimedia. Association for Computing Machinery. doi:10.1145/3746027.3756875. Retrieved 24 June 2026.
- ↑ 4.0 4.1 4.2 "PySimPace v2.0: An Easy-to-Use Simulation Tool with Machine Learning Pipelines for Realistic MRI Motion Artifact Generation". Open Research Exeter. University of Exeter. Retrieved 24 June 2026.
- ↑ "Accepted Papers - Open Source Software". ACM Multimedia 2025. Retrieved 24 June 2026.
External links
- Official website
- PySimPace on PyPI
- PySimPace ACM Digital Library record
- Open Research Exeter record
- ACM Multimedia 2025 accepted open-source software papers
Category:Python software Category:Free software programmed in Python Category:Medical imaging software Category:Magnetic resonance imaging Category:Neuroimaging software Category:Open-source software Category:Scientific software
This article "PySimPace" is from Wikipedia. The list of its authors can be seen in its historical and/or the page Edithistory:PySimPace. Articles copied from Draft Namespace on Wikipedia could be seen on the Draft Namespace of Wikipedia and not main one.
