spant User Manual

Author

Martin Wilson

Published

December 10, 2024

Welcome

This book is a user-guide for performing basic MR spectroscopy analyses with spant. The focus is on single-voxel 1H MRS data acquired from the human brain - and no knowledge of R or programming is assumed. We focus on the use of a standardised analysis pipeline, developed to cover the majority of use-cases, with minimal need for adjusting the default settings.

At its core, spant is a toolkit of MRS processing and visualisation methods and is therefore capable of numerous analyses types, such as functional-MRS (fMRS), J-edited MRS, 31P MRS and MR spectroscopic imaging (MRSI). However, at the time of writing, pipelines for these less common types of analyses need to constructed by the end-user, which will most likely require a basic knowledge of R programming and MRS acquisition methodology. Help to get started with bespoke pipeline development is available in the on-line package documentation. Whilst there are long term plans to support more analysis types, the current priority is single-voxel 1H MRS data acquired from the human brain, with an emphasis on commercially available sequences and from Siemens, Philips and GE.

Acknowledgements

Writing this user guide began in 2024, and it will continue to evolve, keeping up with the best practices for MRS analysis, and cover new pipelines added to the underlying package. Suggestions for changes or additions to the text and references are very welcome and useful contributions will be acknowledged below. These can be as small as pointing out a typo or a missed reference, or as large as adding a new section, but please get in touch first if you’re planning to contribute anything more than a couple of sentences.

  • Thanks to Paul Mullins and Chuck Gasparovic for periodically (and patiently) re-explaining the nuances of water concentration scaling. All errors are my own regarding assumptions and implementation related to water concentration scaling.

  • Thanks to Jon Clayden for developing and supporting numerous R packages related to neuroimaging that spant depends on.