Release Notes

Important Note

All 3.x releases of the viewer require the XNAT 1.8 generation of the main Web app. More specific version requirements are listed below and summarised in the compatibility matrix.. See XNAT OHIF Plugin Compatibility Matrix

v3.7.0

v3.7.0 requires XNAT 1.8.10 or later. It will not run on earlier versions of XNAT.

Version 3.7.0 of the viewer introduces the support for displaying microscopy scans (DICOM Whole Slide Imaging). Due to the large data size of microscopy images (pre-computed tiles in multi-layered resolutions), it has necessitated the implementation of a custom support for the DICOMweb QIDO and WADO services.

New Features

Custom implementation of the DICOMweb QIDO and WADO services as endpoints within the viewer XAPI.
Slide information panel providing view of the slide label image (when available), and switching the visibility between provided channels (optical paths).
Partial support for images with multiple focal planes.
- If provided, the extended-depth-of-field (focus stacking) images are displayed.
- Otherwise, the viewer will try to display the first focal plane.
Upload snapshot images to XNAT session resources.

Fixes

Fixed the freehand3DTool becoming unresponsive when deleting a contour while drawing.
Fixed dependencies related to Node.js modules with Webpack 4.

v3.6.3

Fixed image position sync for scans with different slice thickness.

v3.6.2

Search both DICOM and secondary catalogs for SOP instance UIDs for ROI collection dependency checks.
Support 4-D datasets with duplicate InstanceNumber values.
Additional dimension tags for the grouping of stacks in 4-D datasets.
Dynamic window width calculation when the value is zero.
Fixed issues preventing from parsing and displaying NM images.
Fixed the detection of orientation upon importing segmentations.
Fixed the middle frame calculation used for displaying the thumbnail images.
Fixed irregular spacing wrongly detected as missing frames.

v3.6.1

Automated the detection of 4-D stack dimensions when creating multi-stack displays.
Enabled the grouping of DICOM Enhanced images belonging to the same scan into a single "Cluster".
- A scan may comprise multiple Enhanced images, such as in the case of Siemens XA30 EPI scans. The viewer used to handle such scans by creating individual thumbnail for each image in the Scan panel, which required pulling all the instances from the backend and eventually causing out-of-memory errors. As a temporary solution, the viewer now groups such images into a cluster with a single thumbnail entry in the Scan panel. This configuration will be replaced with a 4-D multi-stack representation in future releases of the viewer.
Display the number of study scans in the Scans panel.
Simplified the data validation for enabling the 3D MPR mode.
Fixed invalid VOI Attributes provided by the cornerstone-wado-image-loader.

v3.6.0

v3.6.0 requires XNAT 1.8.7 or later. It will not install or run on earlier versions of XNAT.

Version 3.6 of the viewer brings new exciting features such as organising and displaying multi-stack images into individual stacks, a dedicated viewport menu to control display options, and image fusion in the 2D viewing mode. Various improvements and bug fixes have also been introduced.

New Features

Organise and display multi-stack frames into subgroups. Only MR scans are currently supported.
Viewport options menu to declutter the display and reduce visual search and selection time.
Support 2D image fusion using cornerstone layers.
Toggle visibility of annotations (masks, contours & measurements) for individual viewports.

Improved

Notify the user of metadata inconsistencies found in individual images.
Added the capability to use false color images instead of colormaps in 2D image fusion.
Make the dialog display optional for new measurements.
Display fusion description and colormap name on the viewport.

Fixes

Remove the polyfill support for IE 11 browser.
Fixed line intersection checks for the reference lines tool.

v3.5.3

Fixed a bug that breaks the app when rendering overlay data over thumbnail images.

v3.5.2

Fixed issue 44: Ensure ROICollections return a non-null, non-empty string to use as a filename.

v3.5.1

Improved

Added Viewer Config API (/viewerConfig). Currently, serves endpoints only to GET and PUT ROI Presets.
Select ROI labels using a list populated from presets stored on XNAT at project-level.
Display overlays in the scan thumbnail view for scans with overlay data.
Deactivated the export of ROI Collections in cases of unsupported modalities or insufficient permission.
Improved alert messages to clarify that viewer will automatically open, surface errors to user if metadata cannot be generated.
Added scrollbar to the scan browser.

Fixes

Fixed a bug in obtaining XNAT user details for guest access.
Fixed a bug preventing guests from loading the viewer.
Fixed bug blocking json generation from subject page "View Images" link.

v3.5.0

v3.5.0 requires XNAT 1.8.7 or later. It will not install or run on earlier versions of XNAT.

Version 3.5 of the viewer features performance improvements to the session metadata creation, serialisation and retrieval. The 3D MPR mode now supports rendering and viewing of contour ROIs, via the creation of surfaces that can be cut through in arbitrary planes. Other features, improvements and bug fixes have also been implemented.

New Features

Improved session JSON creation, serialisation and retrieval processes:
- JSON metadata minification with up to 50% size reduction.
- Optimised CPU and memory utilisation.
ROI contour rendering in the 3D MPR mode.
Volume calculation and display for contour and mask ROIs.
Calculation and display of 2D statistics of mask ROIs to make the behaviour more consistent with that for contour ROIs.

Improved

Populate DICOM Equipment module values from the reference scan for storing ROI Collections.
Display scan modality in the scan browser.
Upgraded underlying JavaScript packages and tools.

Fixes

Fixed measurement import/export issue for multiframe images.
Fixed issues caused when using Node.js v16.14.0.
Restrict NM image fusion to be enabled only for image type 'RECON TOMO'.

v3.4.1

Fixes

Fixed import and export issues of measurement data for DICOM multiframe images.
Set values of the DICOM General Equipment tags based on the viewer attributes: manufacturer, model-name and software-versions.

v3.4.0

Two major new features have been introduced in this XNAT OHIF Viewer 3.4.0 release. A Measurement Service to manage and interact with measurement annotations via a dedicated panel, as well as the ability to exchange measurement collections data with XNAT. The integration of MONAI Label as a new menu within the Masks panel which offers interactive image annotation and segmentation.

New Features

A new Measurement Service to manage and interact with measurement annotations, offering:
- Measurement panel to group measurements into working and imported collections.
- Export and import menus to exchange measurement collections with the XNAT archive.
- Individual measurement interaction buttons: edit metadata, toggle visibility, jump to slice, and remove.
- Measurement API to interface with measurements from other Viewer extensions.
- Support to set and restore presentation state for individual measurements.
An integration client for MONAI Label to facilitate interactive medical image annotation using both interactive and automated annotation models.
Ability to switch between NVIDIA AIAA or MONAILabel tools based on the backend configuration.

Improved

New settings option to load a scan from the middle rather than from the top slice.
Added a progress indicator to monitor the loading progress of images. Currently, does not support multiframe images or images that were removed from cache.

Fixes

Fixed issues caused when switching between the 3D MPR and the standard viewer modes.
Other minor tweaks and bug fixes.

v3.3.0

Version 3.3.0 of the XNAT OHIF Viewer focuses on optimizing the data import process of the Contour-based ROIs. The lazy loading feature allows for quicker and more efficient navigation of contour ROIs, where only required ROIs can be loaded interactively. The concurrent loading of contour ROIs employs web workers to extract the RTStruct polygon data with speed optimization of 65%. The display colors of contour ROIs can now be switched between three different schemes (project template, from metadata, or custom values), where the project color template can be added/edited via the OHIF Viewer ROI Color API.

The mask drawing tools were refactored to enable better interaction with touch screen devices.

New Features

Lazy loading of contour ROIs. Feature is inactive by default but can be enabled via Preferences.
Concurrent loading of contour ROIs using web workers.
Server-side storage of ROI color template at project-level. Please refer to the OHIF Viewer ROI Color API in the XNAT Swagger page.
Switch between contour ROI color schemes - project template, from metadata, or custom values.

Improved

Supported drawing/import/export of mask ROIs for the Ultrasound Image Storage 1.2.840.10008.5.1.4.1.1.6.1.
Refactored segmentation and mask tools for touch-based interaction.
Added UI controls to sort the contour ROI list by name. Applies only to the locked collections.
Used automatically generated notation to identify scans with duplicate series number.
Switched to SVGR loader to enable custom title for icons.
Removed unnecessary ITK modules to reduce the bundle size.

Fixes

Rebuild JSON metadata on certain events if project anonymization enabled and add support for a subject-level event.
Fixed contour ROI import/export for multiframe images.
Fixed the rendering position of the contour sculpt tool.
Other minor tweaks and bug fixes.

v3.2.0

The XNAT OHIF Viewer v3.2.0 introduces new features such as 3D image fusion, and viewer links in the XNAT search listings.

New Features

Multi-volume image fusion in 3D MPR mode.
Two new tools: reference lines and crosshairs.
Importing ROI collections via clicking on relevant icons in the study browser.
Link to open viewer from XNAT search listings with admin configuration for default project listings.

Improved

Improved support for DICOM Nuclear Medicine (NM) IOD.
Orientation marker in the 3D MPR mode.
Keyboard shortcuts for mask undo/redo.
Refactored the mask settings menu for improved interaction.
Upgraded the image loader to use WebAssembly for data decoding.
Display subject’s ID/name in the Viewer’s browser tab.
Cache ROI collection numbers displayed in the Scan Browser to reduce repeated API calls.
Display windowing information in the viewport overlay.

Fixes

Prevent regenerating session JSON on project sharing events.
Fixed mask import for coplanar and perpendicular segmentation data.
Fixed duplicate StudyInstanceUID issue.
Upgraded dependencies to recent applicable versions.
Fixed missing person and user details in AIM metadata.
Fixed image thumbnail re-rendering issue.
Fixed intensity scaling for images with variable window/level values.
Refactored smooth and sync components to maintain state during switching between scans.

v3.1.0

Version 3.1.0 of the XNAT OHIF Viewer brings new features such as the DICOM Tag Browser, segmentation tools (correction scissors & undo/redo), and warning notifications about inconsistencies in scan data. The Scan Browser panel has been refactored to highlight the active scan and show the number of ROI collections available per scan.

The integrated NVIDIA AIAA Client now supports the 3D Deepgrow model, and accurately aligns annotation data received from the AIAA server.

New & Improved

Upgraded to the viewer package v4.9.20 of the mainstream OHIF Viewer (@ohif/viewer@4.9.20).
Filter ROI collection import list based on session label and scan number.
DICOM Tag Browser added, based on a refactored extension from OHIF Viewer. The full list of elements are only displayed for loaded images.
Visual notifications added to warn about inconsistencies in scan data, e.g. frames have different dimensions.
Highlight active scan in the Scan Browser panel, and show the number of available ROIs per scan.
Added new segmentation tools: correction scissors and brush eraser.
Toolbar buttons added to undo/redo manual segmentation.
Support changing segmentation brush size when using touch screen devices.
Smart & Auto segmentation brush: custom gate-separation value + retain the settings between viewer sessions.
Show the number of slices for each segment and jump to the middle slice when value clicked.
Toggle visibility of contour collections.
Added support for 3D Deepgrow in the NVIDIA AIAA Client (requires Clara v4.0+).

Fixes

Display label for non-ROI-collection assessors in the assessment table (Issue #28).
Refactored a new class to read and align NIfTI segmentation data. Added support for cornerstonejs/nifti-image-loader.
Fixed buffer size for reading NIfTI segmentation data.
Fixed DICOM overlays not displaying issue.
Handle errors when loading images and show notification with error message (Issue #22).
Notify about ROI export issues.
Added a progress indicator for importing contour collections.
Fixed a bug that prevented changing Window/Level preferences.
Fixed slice order in image sets.
Fixed the annotations ‘delete all’ function to remove currently displayed measurements.
Fixed segmentation export issue when having multiple viewports.
Fixed Ultrasound measurements to show values in physical units. Fix does not support SequenceOfUltrasoundRegions with a multiple entries.
Fixed imageId issue that prevented scrolling through some multi-frame images.

v3.0.1

Fixes

Fixed JSON metadata not being updated when new data is added to existing session. This was causing viewer to only display part of the session.
Fixed scan date issue.
Added milliseconds to the ROI assessor labels.

v3.0.0

Version 3.0.0 of the XNAT OHIF Viewer focuses on defining regions-of-interest, with the "ROI Collection" datatype now a part of core XNAT, which removes the dependency on having a separate datatype plugin. There are new mask creation tools and an integration of NVIDIA Clara's AI-Assisted Annotation, which seamlessly applies machine-learning models to auto-segment ROIs that are both editable within the viewer and directly save-able within XNAT. There is support for DICOM "fractional segmentations", allowing the representation of volumetric masks that are probabilistic (e.g., allowing the representation of the degree of certainty of machine learning models). ROI creation on 2-D image modalities is also now available. There are additional enhancements to support for contour datatypes

We track core upgrades to the architecture of the main OHIF project, in particular the transition to a more modern React JavaScript framework to improve viewer performance in XNAT and allow the inclusion of major new functionality, such as a 3D-MPR mode for the first time.

New & Improved

Upgraded to the React-based OHIF Viewer.
Integration of the NVIDIA AIAA tools. (Experimental feature available only for sites able to "spin up" server running NVIDIA Clara AIAA.)
Additional masking tools such as Freehand Scissors and Spherical Brush.
Context menu support for contours allowing to copy/paste contrours. (Experimental)
The xnat-roi-plugin is now merged into ohif-xnat-viewer-plugin to roll all functionality into a single jar file.
Improved design of Contour and Segmentation panels
Ability to add contours to 2-D imaging modalities
Support for "fractional" DICOM-SEGs
3-D MPR view with rotatable cross-hairs to allow oblique slicing
Using relative URLs in JSON metadata allows XNAT's host name to change without requiring a metadata regeneration.
Other plugins such as XSync can now use the ROICollection data type schema (integrated into xnat-data-models).
Finer grained JSON metadata regeneration XAPI calls added.
Added xnat-ohif-viewer repo as submodule to allow single command build of whole plugin.

Fixes

Fix event-triggered JSON metadata generation for session uploads.
JSON metadata format update to support OHIF 4.2.7 requirements.