Documentation: first paper draft for circulation

spm · Jan 13, 2025 · 98223d6 · 98223d6
1 parent 42c525e
commit 98223d6
Show file tree

Hide file tree

Showing 2 changed files with 24 additions and 19 deletions.
diff --git a/paper/paper.bib b/paper/paper.bib
@@ -168,13 +168,14 @@ @article{boto2018moving
   publisher={Nature Publishing Group UK London}
 }
 
-@article{o2023combining,
-  title={Combining video telemetry and wearable MEG for naturalistic imaging},
-  author={O’Neill, George C and Seymour, Robert A and Mellor, Stephanie and Alexander, Nicholas and Tierney, Tim M and Bernachot, L{\'e}a and Hnazaee, Mansoureh Fahimi and Spedden, Meaghan E and Timms, Ryan C and Bush, Daniel and others},
-  journal={bioRxiv},
-  pages={2023--08},
+@article{mellor2023real,
+  title={Real-time, model-based magnetic field correction for moving, wearable MEG},
+  author={Mellor, Stephanie and Tierney, Tim M and Seymour, Robert A and Timms, Ryan C and O'Neill, George C and Alexander, Nicholas and Spedden, Meaghan E and Payne, Heather and Barnes, Gareth R},
+  journal={NeuroImage},
+  volume={278},
+  pages={120252},
   year={2023},
-  publisher={Cold Spring Harbor Laboratory}
+  publisher={Elsevier}
 }
 
 @article{spedden2024towards,

diff --git a/paper/paper.md b/paper/paper.md
@@ -9,10 +9,12 @@ tags:
   - EEG
   - PET
 authors:
-  - name: John Ashburner
+  - name: Tim M. Tierney
     affiliation: 1	
   - name: Nicholas Alexander
     affiliation: 1	
+  - name: John Ashburner
+    affiliation: 1	
   - name: Nicole Labra Avila
     affiliation: 1	
   - name: Yaël Balbastre
@@ -45,8 +47,6 @@ authors:
     affiliation: 1	
   - name: Adeel Razi
     affiliation: 1	
-  - name: Tim Tierney
-    affiliation: 1	
   - name: Ryan Timms
     affiliation: 1	
   - name: Peter Zeidman
@@ -66,7 +66,7 @@ bibliography: paper.bib
 
 # Summary
 
-Statistical Parametric Mapping (SPM) is an integrated set of methods for testing hypotheses about the brain's structure and function, using data from medical imaging devices. These are methods are implemented in an open source software package, `SPM`, which has been in continuous development for more than 30 years by an international community of developers. This paper reports the release of `SPM 25.01`, a major new version of the software that incorporates novel analysis methods, optimisations of existing methods, as well as improved practices for open science and software development.
+Statistical Parametric Mapping (SPM) is an integrated set of methods for testing hypotheses about the brain's structure and function, using data from medical imaging devices. These methods are implemented in an open source software package, `SPM`, which has been in continuous development for more than 30 years by an international community of developers. This paper reports the release of `SPM 25.01`, a major new version of the software that incorporates novel analysis methods, optimisations of existing methods, as well as improved practices for open science and software development.
 
 # Statement of need
 
@@ -80,21 +80,21 @@ Statistical Parametric Mapping (SPM) is an integrated set of methods for testing
 - Dynamic Causal Modelling (DCM) for state-space modelling using variational Bayesian methods [@friston2003dynamic].
 - Source localisation for M/EEG data using variational Bayesian methods [@phillips2005empirical].
 
-What unifies these methods are a series of core principles: the use of generative models that explain data (rather than simply describing it), the well-motivated application of parametric statistics for testing hypotheses, and open science practices that have meant that `SPM` has been freely available since its inception.
+These methods share the key principles of employing generative models, parametric statistics and open science practices.
 
 This major new release of `SPM` addresses a series of needs in the neuroimaging community, set out below.
 
 ## Open development
 
-`SPM` was previously developed and tested using a private Subversion server within University College London. To enable community engagement in the future development of `SPM` and to increase transparency, development has recently moved to a public [Github repository](https://github.com/spm/spm). `SPM 25` is the first release of the software following the move to Github. The key advantages of using Github thus far have been:
+`SPM` was previously developed and tested using a private Subversion server within University College London. To enable community engagement in the future development of `SPM` and to increase transparency, development has recently moved to a public [Github repository](https://github.com/spm/spm). `SPM 25.01` is the first release of the software following the move to Github. The key advantages of using Github thus far have been:
 
 - Introducing automated unit and regression tests across platforms.
 - Automating the build process to conveniently generate and release source code and compiled versions.
 - Issue tracking and distributing tasks among developers.
 
 ## Documentation and training
 
-The documentation for `SPM` was previously spread across multiple locations, most of which could not be edited by the community. `SPM 25` is accompanied by a new [documentation website]((https://www.fil.ion.ucl.ac.uk/spm/docs/)), the source code for which is hosted in a public [Github repository](https://github.com/spm/spm-docs). The website has step-by-step tutorials on all of SPM's main features, as well as freely available video recordings of lectures from previous SPM courses covering the mathematical theory.
+The documentation for `SPM` was previously spread across multiple locations, most of which could not be edited by the community. `SPM 25.01` is accompanied by a new [documentation website](https://www.fil.ion.ucl.ac.uk/spm/docs/), the source code for which is hosted in a public [Github repository](https://github.com/spm/spm-docs). The new website has step-by-step tutorials on all of SPM's main features, as well as freely available video recordings of lectures from previous SPM courses covering the mathematical theory.
 
 ## Major new features 
 
@@ -108,7 +108,7 @@ The documentation for `SPM` was previously spread across multiple locations, mos
 
 ### M/EEG
 
-- Methods for spectral decomposition - `SPM 25` offers an implementation of an existing approach called FOOOF (Specparam) in the MEEGtools toolbox, based on code from Brainstorm [@donoghue2020parameterizing], as well as a new Bayesian implementation that introduces formal statistical testing, called Bayesian Spectral Decomposition (BSD) [@medrano2024bsd].
+- Methods for spectral decomposition - `SPM 25.01` offers an implementation of an existing approach called FOOOF (Specparam) in the MEEGtools toolbox, based on code from Brainstorm [@donoghue2020parameterizing], as well as a new Bayesian implementation that introduces formal statistical testing, called Bayesian Spectral Decomposition (BSD) [@medrano2024bsd].
 
 - Support for fusion of different MEG sensor types and EEG sensors in beamforming with pre-whitening [@westner2022unified].
 
@@ -118,11 +118,11 @@ The documentation for `SPM` was previously spread across multiple locations, mos
 
 - Parametric Empirical Bayes (PEB) [@friston2016bayesian] extends the Dynamic Causal Modelling (DCM) framework to include random effects modelling of neural connectivity parameters, enabling people to test hypotheses about the similarities and differences among research participants.
 
-- Bayesian model reduction (BMR) [@friston2018bayesian] enables the statistical evidence to be rapidly scored for large numbers of competing models, where models differ only in the priors.
+- Bayesian model reduction (BMR) [@friston2018bayesian] enables statistical evidence to be rapidly scored for large numbers of competing models, where models differ only in their priors.
 
 ### OPMs
 
-A major recent innovation in neuroimaging is MEG using Optically Pumped Magnetometers (OPMs), which enable free movement of the head and body during neural recordings [@boto2018moving]. This makes MEG available to new experimental paradigms (e.g., experiments involving free movement [@o2023combining]), new body parts (e.g., the spinal cord [@spedden2024towards]) and new study populations who may not be amenable to traditional MEG (e.g., people with epilepsy [@mellor2024detection]). Developing analysis tools for OPM data is a major focus for SPM, with recently added features including:
+A major recent innovation in neuroimaging is MEG using Optically Pumped Magnetometers (OPMs), which enable free movement of the head and body during neural recordings [@boto2018moving]. This makes MEG available to new experimental paradigms (e.g., experiments involving free movement [@mellor2023real]), new body parts (e.g., the spinal cord [@spedden2024towards]) and new study populations who may not be amenable to traditional MEG (e.g., people with epilepsy [@mellor2024detection]). Developing analysis tools for OPM data is a major focus for SPM, with recently added features including:
 
 - File IO for all major OPM manufacturers (Quspin, Cerca, Mag4Health, Fieldline).
 - Methods to simulate arbitrary OPM arrays of differing densities and vector measurements.
@@ -131,7 +131,7 @@ A major recent innovation in neuroimaging is MEG using Optically Pumped Magnetom
 
 ### SPM without MATLAB
 
-Approximately 90% of the `SPM 25` source code is written in MATLAB and the remainder is C++. This code has been highly optimised and thoroughly tested over 30 years of development. We have therefore carefully considered how to capitalise on the stability of the `SPM` software, while making it more  accessible for people who do not have access to a MATLAB license, or who prefer to write their analysis code in other languages. 
+Approximately 90% of the `SPM 25.01` source code is written in MATLAB and the remainder is C++. This code has been highly optimised and thoroughly tested over 30 years of development. We have therefore carefully considered how to capitalise on the stability of the `SPM` software, while making it more  accessible for people who do not have access to a MATLAB license, or who prefer to write their analysis code in other languages. 
 
 Our strategy is as follows:
 
@@ -141,9 +141,13 @@ Our strategy is as follows:
 
 - [Docker and Singularity containers](https://www.fil.ion.ucl.ac.uk/spm/docs/installation/containers/) are additionally provided and are now generated automatically as part of SPM's Github build process.
 
+# Software versions
+
+`SPM 25.01` is the first release of `SPM` to use calendar versioning, thus `SPM 25.01` is the version issued in January 2025. All releases are available via [https://github.com/spm/spm/releases](https://github.com/spm/spm/releases). 
+
 # Acknowledgements
 
-JM is supported by the Discovery Research Platform for Naturalistic Neuroimaging funded by Wellcome
-[226793/Z/22/Z]. PZ is funded by an MRC Career Development Award [MR/X020274/1].
+Tim M. Tierney is funded by an Epilepsy Research UK fellowship (FY2101). Johan Medrano is supported by the Discovery Research Platform for Naturalistic Neuroimaging funded by Wellcome
+[226793/Z/22/Z]. Peter Zeidman is funded by an MRC Career Development Award [MR/X020274/1]. A full list of authors of `SPM 25.01` can be found in the file `AUTHORS.txt` supplied with the software.
 
 # References