New Features |
|---|
| Data Analysis Management | The new data management tools enable a new way of handling and working with (large) data sets: Instead of isolated documents, data is entered and managed in a hierarchical manner. At the top level, projects are created that refer to the data of a whole experiment. Under each specified project, subjects and workflows are added. For each project, workflows can be defined that process input data and create desired output data. Since workflows know about the data belonging to a project, processing can be performed in the same way for the data of all subjects providing a powerful batch processing mechanism without the need to write scripts. Furthermore, workflows create detailed reports in HTML and PDF format as a basis for quality assurance and documentation of performed analysis steps. For details about the new data management tools and supported workflows, consult chapter "Data Analysis Management" in the User's Guide. |
| MNI Normalization | This version supports MNI space normalization using automatic template-based alignment. MNI space normalization can be selected as an alternative to Talairach space normaliztion for individual dcouments using the "Volumes > Normalize To MNI Template Space" menu, as well as for normalization workflows in the context of the new data management analysis tools. The VTC creation dialog has been updated to allow specifying a 12 parameter MNI transformation file to create MNI space VTC files. For further details, consult the "Brain Normalization > MNI Normalization" section in the "Basic (f)MRI Data Analysis" chapter of the User's Guide. |
| Representational Similarity Analysis | Representational similarity analysis (RSA) uses activity patterns in fMRI data to analyze the response similarity between conditions in selected regions-of-interest. For each ROI a representational dissimilarity matrix (RDM) is computed and graphically displayed containing distance measures (1-correlation) between pairs of distributed activity patterns. Multi-dimensional scaling (MDS) is performed to visualize the similarity structure in two dimensions. At a second level, the calculated similarity structure between conditions can itself be compared across ROIs. Since data from various sources can be integrated in the analysis, this second-level procdedure allows to compare RDMs across subjects, other measurement modalities as well as between computational models without requiring voxel-level correspondence. For further details, see chapter "Representational Similarity Analysis" in the User's Guide. |
| VTC-VTC Alignment | Analyzing sub-millimeter fMRI data poses specific challenges for the alignment of the data from multiple runs of an experiment within a scanning session since even small misalignments will strongly reduce the detectability of fine-grained feature or layer-specific activation clusters. The alignment of sub-millimeter fMRI runs is also difficult since such experiments usually acquire only a small portion (slab) of the brain. In order to perform optimal across-run alignment a VTC-VTC Grid Search Alignment tool has been introduced that works at the level of, optionally masked, VTC data. The VTC-VTC alignment tool uses spatial cross-correlation to assess how well the the source VTC's first volume and target VTC's first volume, match each other. For further details, see topic "VTC-VTC Alignment of Multiple Runs" in the "Analysis of Sub-Millimeter 7T+Data" chapter of the User's Guide. |
| Open Documents Panel | The main window provides a new "Open Documents" panel that is always available via a toggle icon in the main toolbar. The panel provides a graphical overview of available (open) documents presented in a "cover flow" layout. The Open Docs panel is especially useful when many documents are available since it allows to select a document using its current visual representation. By moving the mouse to the left and right while holding it down it is also possible to quickly browse through the documents and to select a document. Furthermore, a right (context) click on the central (current) document provides a context menu (like for tabs) that can be used to call relevant document-specific functionality. |
| Zoom View Panel | The main window provides a new "Zoom View" panel that is always available via a toggle icon in the main toolbar. Using a mouse selection while holding down the ALT key, any rectangular section of the current FMR or VMR document can be selected to be displayed in the Zoom View panel. This is especially useful for inspecting spatial details of the anatomy and overlaid functional maps from high-resolution (sub-millimeter) scans. For VMR documents, the panel also allows to draw inside the volume using the zoomed section as a proxy to the data. For each open VMR document, the program also keeps a full drawing history allowing undo/redo operations for manual segmentation operations. The zomming rectangle can also be moved to new locations by clicking inside while holding down the ALT key follwoed by mouse movenents. For more details, consult the "Manual Segmentation Tools" topic in the "Brain and Cortex Segmentation" chapter of the User's Guide. |
| Python for Scripting and Plugin Development | Python is an increasingly popular language for scientific programming as well as scripting. In this version, experimental support for Python is provided (for Mac and Windows 64-bit) with the goal to provide an easy and unified alternative to both the current JavaScript scripting as well as C++ plugin development. The core of the new tools forms an embedded Python interpreter (based on PythonQt, [link]). Since the Python interpreter is enriched with access to the Qt API, it is easy to add cross-platform user interfaces for scripts and plugins in Python. The interpreter also has full access to the BrainVoyager API enabling scripting and batch programming in the same way as with the embedded JavaScript environment. Since Python supports powerful modules for scientific computing such as NumPy and SciPy, the same language can also be used to extend BrainVoyager's computational capabilities. Plugin development is also supported by the "bv" Pyton C module that provides access to internal BrainVoyager volume and surface data as numpy arrays. For details, consult section "Additional Documentation - Python Scripting and Development" in the BrainVoyager User's Guide. |
Enhancements |
|---|
| Phase Encoding Based EPI Distortion Correction | The COPE (Correction based onOppositePhaseEncoding) plugin was available as an experimental feature with BVQX v2.8.4 and is now released as version 1.0 after substantial testing and additional improvements. The new release contains also embedded information ("Help" tab) explaining the options of the plugin in detail. |
| Fast FMR Creation for Mosaic Dicoms | Reading and parsing Siemens mosaic DICOM files has been substantially improved. The creation of FMR and DMR documents from such mosaic files will, thus, be much faster than in previous versions. |
| Overlay GLM: Multiple contrasts and contrast names | The "Overlay GLM" dialog now supports overlaying multiple defined contrasts when selecting the new "Overlay all contrasts" option. A special set of contrasts, where each main predictor gets a 1 contrast weight and all other predictors a 0 weight, can now be created by using the "Add 1 Per Pred" button. The dialog now also creates better default contrast names for resulting maps that are based on the involved betas when selecting the new "Include predictor names in contrast name" option (on as default). |
| Mesh Depth Sampling of Multiple VMPs | When providing a data VMP file in the "Cortical Depth Sampling" dialog, only the first volume map was used to create surface maps at the specified depth levels. The tool now creates depth-level surface maps for all maps in the provided data VMP file when using the "Create depth surface maps" option is turned on. |
| 12 Parameter Transformation Matrices | The new MNI transformations use matrices with 12-parameters describing shear values in addition to the previously supported translation, rotation, scale values. These matrices are fully supported and also displayed by checking the "Matrix" option in the "Coregistration" tab of the "3D Volume Tools" dialog. |
| ROI-GLN Beta Bar Plots | When running ROI/VOI GLMs, the estimated beta values are now shown in a bar graph. Also the voxel-beta plot now shows bar plots instead of the previously used line plots; line plots can still be used by changing the default plot style in the "Settings" dialog. |
| CBA Improvements | Cortex-based alignment can now be used for non-1mm VMR data sets, including segmented brains from sub-millimeter human and monkey MRI. |
| Curvature calculation of current mesh state | When calculating mesh curvature maps, the curvature of the linked mesh of another folded state was used if established. While this is often useful, it has been cumbersome to calculate curvature for the current state of the mesh, e.g. when smoothing or inflating a mesh; it required unlinking a mesh in the "Mesh Morhphing" dialog first, which then needed to be often re-established. The new Use "linked mesh" option in the "Background and Curvature Colors" dialog can now be used to decide for which mesh state the curvature should be calculated. |
| FMR 3D ROIs | ROIs of FMR projects are now true 3D ROIs. This increases compatibility with the real-time fMRI Turbo-BrainVoyager software. |
| Target Space for VMR Transformation | When spatially transform VMR data sets using the "Spatial Transformation of VMR" dialog, one could edit the name for the resulting VMR but the reference space was set to "unknown". The dialog now has an option to set the target space (Unknown, Native, ACPC, MNI), which will also modify the file name accordingly. |
| New Scrpting Commands | Note that for Python scripting, the "BrainVoyager" object must be used instead of the "BrainVoyagerQX" object for JavaScript scrpting (that will also transition ot "BrainVoyager" in the future). It is now possible to ask for the number of volume and surface maps using the "NrOfVolumeMaps" and "NumberOfSurfaceMaps" functions; Python scripts/plugins can also access the content of maps using the "bv" extension module. Processed fuctional (FMR) data can be saved to disk using the "SaveFMRAndSTCFromMem" command. The "MeshScene" property of a VMR document provides access to an existing surface window; in case that a surface window should be created when not yet existing, the "GetMeshScene" function must be used. When updating the visualization of a mesh object after changing curvature or other maps, the "UpdateAppearance" command can be used (the "UpdateSurfaceWindow" command of the hosting VMR doucment is not sufficent since it does not update surface maps). The "UpdateWindow" command of VMR objects will update also the state of visualized volume maps. Volume and surface maps can be loaded using the "LoadVolumeMaps(filename)" and "LoadSurfaceMaps(filename)" script commands and a specific VMP or SMP sub-map can be displayed using the "ShowVolumeMap(index)" or "ShowSurfaceMap(index)" command for VMR and mesh objects, respectively; furthermore, the names of a specific volume or surface sub-map can be extracted using the "GetNameOfVolumeMap(index)" and "GetNameOfVolumeMap(index)" commands. |
Bug Fixes |
|---|
| ACPC-TAL for sub-millimeter VMRs | The ACPC to TAL and Un-TAL functions did not properly work for high-resolution (sub-millimeter) VMRs with large bounding boxes (384, 512 or higher). This issue has been fixed. Large VMRs are now also supported by GPU sinc interpolation. |
| Brdge removal for sub-millimeter VMRs | The bridge removal tool now supports non-1mm resolution non-256 bounding box VMRs without removing information about bounding box, voxel resolution and other extended-header information. |
| Consistent cluster colors | When converting volume map clusters into surface clusters, the color of sub-clusters within a map did not all received the correct color (soe got the default blue color). This issue has been fixed. |