This page lists the bugs, bug fixes, and minor changes to Jim version 5.0. Any minor changes should not affect core functionality of Jim unless otherwise indicated. Fixed versions of Jim have a minor release indicated by a number after the underscore "_", such as "5.0_1". You can see which version you currently have installed by selecting About from the Help menu. Comparison with the table below will indicate whether a bug you have noted needs to be reported, or has already been fixed.
If you're looking for the latest bugs and fixes for Version 4.0, you need to look at the Jim 4.0 bug parade. If you're looking for the latest bugs and fixes for Version 3.0, you need to look at the Jim 3.0 bug parade.
If you want to install the latest release of Jim 5.0, you can either do a full re-install, or you can simply use this link to download the file xinapse5.jar. Click right on the link above, and save to a temporary location before using it to replace the file called xinapse5.jar in the install directory. If you don't know which is the install directory, do:
Important Note: if you do a full re-install of the latest version of Jim 5.0, and have performed any customisation of the installation (such as editing the Unix shell script Jim), then please be sure and make backup copies of these changed files in another directory before installing Jim again. If your license.txt file is in the Jim installation directory, then please also ensure that you make a backup copy of license.txt.
Key: type of issue
| Type | Bug/Enhancement Description | Fix Description | Bug ID | Status | Version / date |
|---|---|---|---|---|---|
| b | Java stack trace shown with NullPointerException thrown when loading NIFTI images with UNKNOWN extended header data. | Fixed. | 50-075 | F | 5.0_33, ?? Apr 2016 |
| b |
Stack trace shown when launching Resampler or Masker on certain Linux platforms
(CentOS, Ubuntu etc.)
|
Fixed. | 50-074 | F | 5.0_32, 09 Jan 2014 |
| b | Update to Bug 50-073. Keyboard accelerators stop working after interaction with ROIs. Seems to affect only certain distributions of Linux. | Again, potentially fixed, but please report any further problems to support "at" xinapse.com. | 50-073 | F | 5.0_31, 23 Jun 2011 |
| b | Update to Bug 50-072. Keyboard accelerators stop working after interaction with ROIs. | Potentially fixed, but please report any further problems to support "at" xinapse.com. | 50-073 | F | 5.0_30, 19 May 2011 |
| e | Added two extra CT contrast presets for mediastinum and abdominal soft tissue, at the request of a user. | - | - | - | 5.0_30, 19 May 2011 |
| b | Keyboard accelerators stop working after interaction with ROIs. | Potentially fixed, but please report any further problems to support "at" xinapse.com. | 50-073 | F | 5.0_29, 10 May 2011 |
| e | Interleaver now enables inter-slice spacing (pixel-z-size) to be the same as the first input image, rather than being equal to the pixel-z-size of the first input image divided by the number of images interleaved. | This option is to allow the interleaving of images with different contrasts rather than images with gaps between slices. | - | - | 5.0_28, 02 Jan 2011 |
| b | BrainFinder tool fails to calculate brain parenchymal fraction (BPF) on some platforms. Jim states that BPF is done, but no BPF result pops up. | Fixed. Timing bug. | 50-072 | F | 5.0_28, 02 Jan 2011 |
| b | SliceExtractor, Concatenator and Interleaver do not work correctly with NTFTI-1 images in one-file format (.nii). They can produce output images that cannot be loaded into Jim when the "Save to disk" option is selected. | These tools placed pixel data on disk, before ExtendedData was added to the NIFTI-1 images. When pixel data is already on disk, ExtendedData cannot be added to a NIFTI-1 image. These tools now add ExtendedData to NIFTI-1 images before pixel data is written. | 50-071 | F | 5.0_27, 2 Nov 2010 |
| b | Conversion of Bruker ParaVision™ images fails on Windows platform. | Jim made assumptions about file naming and path separators that were Unix-specific and do not apply to the Windows file naming. | 50-070 | F | 5.0_27, 2 Nov 2010 |
| b | ROIs not sent on DICOM send to remote DICOM server. | With large ROI files, ROIs are not correctly sent to the server. | 50-069 | F | 5.0_27, 2 Nov 2010 |
| b | DICOM Query/Retrieve does not work with some DICOM servers. | Fixed. Caused by non-standard C-FIND response from one particular type of server. Fix involves a work-around for the non-standard response. | 50-068 | F | 5.0_27, 2 Nov 2010 |
| B | Further update to Bug ID 50-064. Bruker ParaVision™ images not converted correctly where in-plane pixels sizes are not equal size and where read and phase-encoded directions are transposed in the image data. | Fixed. Caused by failure to transpose pixel sizes and image dimensions. | 50-067 | F | 5.0_26, 15 Sep 2010 |
| b | Incorrect handling of explicit-VR DICOM elements that were were of type "Other Float" and "Unlimited Text". This bug became apparent when attempting to load a DICOMDIR file on a CD-ROM produced by an Agfa PACS system, which contained Unlimited Text elements. | Fixed. | 50-066 | F | 5.0_26, 15 Sep 2010 |
| B | Pre-processing registration performed in Perfusion and DCE-MRI toolkits does not correct mis-registration when image mis-registrations are small. Correction of in-plane rotations also not performed correctly. | Fixed. | 50-065 | F | 5.0_25, 07 July 2010 |
| B | Bruker ParaVision™ images not converted correctly for 3-D pulse sequences; when image matrix is non-square; or where in-plane pixels sizes are not equal. | Fixed. Caused by reversal of pixel size and matrix values in the row and column directions. | 50-064 | F | 5.0_25, 07 July 2010 |
| b |
Jim can hang (not respond) in any of the following situations:
|
Fixed. Caused by bug in latest Java versions 6.0 update 18, u19 and u20. See: Sun bug ID 6880336. Please report any further problems with hanging, quoting the Java version you use. | 50-063 | F | 5.0_25, 07 July 2010 |
| - | Improved the instructions (for Solaris and Linux systems) on how to configure the "LicenseServer" and "DicomStorage" programs so that they start automatically on system reboot. | - | - | - | 5.0_24, 12 May 2010 |
| - | Scripts for stand-alone tools changed to use the bourne shell, instead of the c-shell to avoid problems with passing arguments between scripts. | - | - | - | 5.0_24, 12 May 2010 |
| B | When Image Converter is used and "Load Result" selected, the displayed image did not necessarily have the correct left-right or top-bottom orientation. | Fixed. Please report any strange discrepancies with image orientation. | 50-062 | F | 5.0_24, 12 May 2010 |
| B | On Mac OS X platform, Analyze 7.5 images not written correctly. Images created by Image Converter (and other tools) have pixel values that are all zero. This bug was introduced in release 5.0_20 (bug ID 50-049). | Fixed. Caused by an out-of-date library javax.vecmath on Mac OS X 10.6 (Snow Leopard). Library functions not present in the old library are now not used. | 50-061 | F | 5.0_24, 12 May 2010 |
| B | Pixel z-size (spacing between pixel centres in slice direction) not set correctly for some imported Varian images. | Fixed. For multi-slice datasets, the location was being read in millimetres, but in reality is in centimetres. | 50-060 | F | 5.0_24, 12 May 2010 |
| B | In Brain Perfusion tool, when automatic AIF finder is used, the AIF found depends on the arterial relaxivity and arterial haematocrit value entered. | Fixed. The automatic AIF found is now independent of arterial relaxivity and arterial haematocrit. The AIF found is now the same as if values entered were 1.0 for arterial relaxivity and 0.45 for arterial haematocrit. | 50-059 | F | 5.0_24, 12 May 2010 |
| b | Incorrect units for relaxivity shown in Perfusion and DCE-MRI toolkits. | Fixed. Units of relaxivity are now shown as L s-1mol-1 (Litres per second per mole). This fix only affects the units label, and does not affect quantification. | 50-058 | F | 5.0_24, 12 May 2010 |
| e | Enhancement to Roaming Fit dialog, when exporting fit graph as PDF document now shows fit variables and fitted values. | - | - | - | 5.0_23, 04 Feb 2010 |
| b | When Image Resampler is used to crop an image, the image crop dimensions were incorrectly recorded in image Audit Trail information. | Fixed. | 50-057 | F | 5.0_23, 04 Feb 2010 |
| b | Image Statistics dialog, when exported as PDF document, does not show mean intensity for the histogram. | Fixed. | 50-056 | F | 5.0_23, 04 Feb 2010 |
| b | When an erase operation is performed on an Irregular or CurvedLine ROI, the resulting ROIs do not all retain the same user colour as the original. | Fixed. | 50-055 | F | 5.0_23, 04 Feb 2010 |
| b | Bad ordering of slices when saving NIFTI-1 images with the new SPM-compatibility option (Interoperability Preferences) set. | Fixed. SPM-compatible image orientation set in NIFTI header, but slices not reversed if slice ordering needed to be reversed to match. | 50-054 | F | 5.0_22, 04 Jan 2010 |
| e | Enhancement to floating license to allow the user to save work when network connection to license server is interrupted. It is possible for Jim to loose a floating (network) license to another computer if the network connection to the License Server is interrupted, and another computer requests a license during the network interruption. | Jim now allows a period of grace after a network connection is lost and restored in the event of losing the license to another computer. A warning is issued, encouraging the user is to save any work-in-progress, and then to restart Jim. | - | F | 5.0_21, 15 Oct 2009 |
| b | Image header information (such as DICOM Elements) were not stored when using the Image Converter to create NIFTI images. | Fixed. The class NIFTIImage did not implement InfoStorer. | 50-053 | F | 5.0_21, 15 Oct 2009 |
| b | Text (Annotation) ROIs are always drawn with the same default colour as other ROIs, even when the default text colour is changed in the ROI user preferences. | Fixed. | 50-052 | F | 5.0_21, 15 Oct 2009 |
| B | When writing ROI statistics, empty statistics files could be created depending on the ROI statistic used to sort the order of the statistics written. | Fixed. | 50-051 | F | 5.0_20, 15 Sep 2009 |
| b | When an image is saved, the default file name set in the FileChooser to select the saved file name may have spurious extensions such as ".hdr.gz.hdr.gz" or ".nii.gz.nii.gz" when working with compressed Analyze/NIFTI images. | Fixed. | 50-050 | F | 5.0_20, 15 Sep 2009 |
| b | Image orientation not set correctly when saving an Analyze 7.5 image with coronal or sagittal orientation using the "Save As" feature in the Jim's main display. Image was always saved with an axial orientation set in the header. | Fixed. | 50-049 | F | 5.0_20, 15 Sep 2009 |
| e | Added new Colour Mapping lookup table "Cool". | - | - | - | 5.0_19, 11 Aug 2009 |
| b |
Problem with creation of one-file compressed NIFTI images when using command-line
driven tools. Tools generate error messages like:
Masker: ERROR: some_mask.nii.gz (No such file or directory). |
Fixed. Xinapse interface to NIFTI-1 file format did not test for existence of all variants of one-file/two-file compressed/uncompressed images after they had been created on disk. | 50-048 | F | 5.0_19, 11 Aug 2009 |
| b |
The following dongle licensing issues have been addressed:
|
Fixed. Dongle licensing is now available for 64-bit Intel Mac OS X systems, and also on 64-bit Linux. Support for dongle licensing on these systems will will need a full re-install of Jim (but not java or java3d). | 50-047 | F | 5.0_19, 11 Aug 2009 |
| e | Added new preset to the Image Fitter tool, for the calculation of R2 images from multi-echo-time images. This complements the preset for calculating T2. | - | - | - | 5.0_18, 15 July 2009 |
| b | Could not convert Bruker ParaVision™ images where the file names were in upper-case. | Fixed. Conversion of Bruker images is now insensitive to the case of the file names. | 50-046 | F | 5.0_18, 15 July 2009 |
| b | Incorrect time zone outputted for Scan_Date image header field in UNC and NIFTI-1 images. The time zone was for the zone in which the image conversion took place, and not the time zone where the images were acquired. | Fixed. Time zone has been removed. | 50-045 | F | 5.0_18, 15 July 2009 |
| e |
Enhancement made to Image Statistics dialog when writing text and exporting PDF versions of
the image statistics.
|
- | - | - | 5.0_17, 18 Jun 2009 |
| b | Java stack trace sometimes shown when loading images with 32-bit integer pixel values. | Fixed. Caused by timing problem when drawing intensity slider labels. | 50-044 | F | 5.0_17, 18 Jun 2009 |
| b | Image intensity histograms incorrectly drawn into PDF reports when the histogram has a very large number of bins, typically seen in images with Integer data types. The histograms could be invisible when the bin width was smaller than one pixel. | Fixed. These histograms are now drawn using anti-aliasing. | 50-043 | F | 5.0_17, 18 Jun 2009 |
| b | Failed to load Analyze 7.5 images with illegal image orientation value. | Fixed. If an Analyze 7.5 image is encountered with an illegal image orientation, the orientation is now assumed to be "axial (flipped)" (orient = 3). | 50-042 | F | 5.0_17, 18 Jun 2009 |
| b | When loading an image by drag 'n' drop, the current loaded image was not unloaded cleanly, so for example, the user was not asked whether she wanted to save ROIs. | Fixed. | 50-041 | F | 5.0_17, 18 Jun 2009 |
| b | BrainFinder fails to run, with error message "you may not set both a manual threshold and a BPF threshold" if a manual threshold between CSF and brain is set (option -t). | Fixed. Incorrect checking for conflicting arguments. | 50-040 | F | 5.0_17, 18 Jun 2009 |
| e | Added an "Extra-High" fidelity setting to the Iso-Surface selector dialog for surface-rendering. This uses approximately 3.3 times more image voxels to generate the surface than the "High" setting, and may be very slow on older computers. | - | 50-039 | F | 5.0_16, 21 May 2009 |
| b | 3-D viewer fails to display rendered surface for ROI Selector when cut-out is set, on some NVIDIA graphics cards, when large images are used. | Image used in the creation of the rendered surface is downsampled so that fewer surface triangles are generated. Please report any further problems with the 3-D display, sending details of your graphics card. | 50-039 | F | 5.0_16, 21 May 2009 |
| b | Display geometry set in User Preferences for main display is ignored. | Fixed. | 50-038 | F | 5.0_16, 21 May 2009 |
| b | When the Orthogonal Views dialog is launched and the loaded image has no orientation information, no default orientation is set the in the Orthogonal Views dialog, and the image plane location cannot be moved until an orientation (Axial, Coronal or Sagittal) is selected. | Fixed. | 50-037 | F | 5.0_16, 21 May 2009 |
| b | Java stack trace shown in Error Console when changing Orthogonal Views position while editing ROIs. | Fixed. An Irregular ROI can be accidentally 'picked up' by the mouse while changing the Orthogonal Views position, resulting in an erroneous attempt to move a vertex. | 50-036 | F | 5.0_16, 21 May 2009 |
| E | Enhancement to Interoperability Preferences in Jim's main Preferences dialog. A new preference has been supplied to reading and writing NIFTI-1 images with no orientation information in a mode which is compatible with fsl. |
With this option is set, when reading a NIFTI-1 image with no orientation information
the image is treated as an axial image with increasing patient left coordinate with
increasing stored pixel data column number.
When writing NIFTI-1 images, even when the orientation is unknown, both the sform qform fields in the header are set to be compatible with an axial image. Use this Preference setting in combination with the SPM-compatibility option to write only images orientations that are compatible with FSL and SPM (i.e., increasing patient left coordinate with increasing stored pixel data column number). |
- | - | 5.0_15, 25 Mar 2009 |
| b |
Java stack trace when in Error Console when:
|
Fixed. Did not check for a non-null Colour Bar dialog in the above circumstances. | 50-035 | F | 5.0_15, 25 Mar 2009 |
| b | Image contrast could not be set for slices individually, even when contrast was not 'auto-linked' between slices. De-selecting a slice caused the contrast setting for that slice to be set the values for the newly-selected slice. | Fixed. | 50-034 | F | 5.0_15, 25 Mar 2009 |
| b | Jim could not handle colour (RGB) with more than 268,435,456 (228) pixels. Now Jim should be able to handle RGB images with up to 715,826,882 pixels, Complex images with up to 1,073,741,824 pixels and all other types of image with up to 2,147,483,647 pixels. | Integer overflow problem, fixed. Please report any further problems when handling large images. | 50-033 | F | 5.0_15, 25 Mar 2009 |
| e | Enhancement to Jim's command-line options to allow multiple images to loaded into spawned image frames. Previously, only one frame could be spawned from the command line; now up to 10 can be spawned, each with its own loaded image and geometry. | For details of these new option, please use the command-line option --help when starting Jim. | - | - | 5.0_14, 06 Mar 2009 |
| E | Enhancement to Interoperability Preferences in Jim's main Preferences dialog. A new preference has been supplied to allow NIFTI-1 images created when reorienting images with the Image Resampler and Multi-Planar Reconstruction to be saved with an orientation that is compatible with SPM and fsl. Inter |
In order for SPM and FSL to display images in a correctly-oriented fashion on screen, they
seem to require pixels to be stored on disk in a particular order with respect to anatomical
coordinates. If you have problems viewing images in SPM and FSL, please try setting the new
SPM-compatibility option in the Interoperability Preferences.
Please report any problems when reorienting images. |
- | - | 5.0_14, 06 Mar 2009 |
| b | When the Orthogonal Views dialog is launched, correct orientation of an image which is already displayed is not set in the dialog. | Fixed. | 50-032 | F | 5.0_14, 06 Mar 2009 |
| b | Reports that some images created by Jim cannot be successfully read by external tools (such as SPM) until Jim is exited. We have been unable to reproduce this, but it may be operating-system dependent. The problem was reported with Jim running under Microsoft Windows. | We have attempted to fix this problem by providing additional synchronization when images are closed, between the internal Java file descriptors and the operating system's file descriptors. If you have this problem please report it, and let us know your operating system. | 50-031 | F | 5.0_14, 06 Mar 2009 |
| b |
In Multi-Planar reconstruction, when changing field-of-view of the resampling volume,
the selected volume does not respond correctly and the orthogonal views are not draw
correctly. In addition, a java stack trace may be shown with an
IndexOutOfBoundsException.
This bug became apparent only when testing under OpenSuSE 11.1. |
Fixed. Timing error. | 50-030 | F | 5.0_14, 06 Mar 2009 |
| b | When Orthogonal Views dialog is shown for the main display and slave displays, setting the cursor type in an Orthogonal Views dialog leads to the cursor type buttons disappearing from other Orthogonal Views dialogs. | Fixed. | 50-029 | F | 5.0_13, 13 Feb 2009 |
| b |
When re-orienting NIFTI-1 and UNC images using either the Image Resampler or
Multi-Planar Reconstruction tools, reoriented/resampled position and
orientation information may not be set correctly.
Furthermore, in NIFTI-1 images, although the qform position/orientation information may be set correctly, the sform information may remain unaltered. Thus, programs that use the sform in preference to the qform to obtain image orientation orientation will read the wrong orientation values. |
Setting of position/orientation information is now done correctly when
resampling/reorienting with ImageResampler and Multi-Planar
Reconstruction. Whenever the positional information is set for a NIFTI-1
image, both the qform and sform are updated.
Please report any further problems when reorienting images. |
50-028 | F | 5.0_13, 13 Feb 2009 |
| b | Incorrect error message in Image Fitter when single input images does not have a number of slices that matches with the number of independent variable values. | Corrected. Message now states that the mismatch is with the number of independent variable values, rather than the number of fit variables. | 50-028 | F | 5.0_13, 13 Feb 2009 |
| e | Added time-to-peak output parameter map for Brain Perfusion toolkit. New map has suffix "TTP", and represents the time (in seconds) from the first appearance of contrast agent in the feeding vessel to the peak of concentration in the tissue. | - | - | - | 5.0_12, 07 Jan 2009 |
| b | For Brain Perfusion and DCE-MRI toolkits, output parameter images do not have image position/orientation information set when input images are in Analyze or NIFTI format. | Fixed. | 50-027 | F | 5.0_12, 07 Jan 2009 |
| b | In the Brain Perfusion toolkit, once either the "Apply smoothing filter", or "Use Brain Finder" options is selected both options cannot be deselected together. | Fixed. | 50-026 | F | 5.0_12, 07 Jan 2009 |
| b | When loading NIFTI files with .nii.gz filename extensions into a display, and then subsequently using the File/Reload function, Jim tries to reload the file without the .gz extension and fails. | Fixed. | 50-025 | F | 5.0_12, 07 Jan 2009 |
| b | Java stack trace may be shown in Error Console dialog when attempting to mask images with ROIs that have been reloaded from disk. This occurs when a loaded Irregular ROI has zero area. An Irregular ROI may have zero area if an ROI is very small, and rounding errors cause adjacent point to be identical when the ROI is saved to a disk file. | Fixed. When tiny, zero-area IrregularROIs like this are now reloaded from disk, they are not added to the current set of ROIs. | 50-024 | F | 5.0_11, 11 Dec 2008 |
| b |
When using "Roaming Response" in Perfusion, DCE-MRI and Time Series
Registration tools, when all input images are not of the same data type, a
stack trace like the following may be seen.
|
Jim has been modified to allow images with mixed data types to be used in these tools. Please report any further problems when working with images of mixed data types. | 50-023 | F | 5.0_11, 11 Dec 2008 |
| b |
Problems loading floating-point images which contain pixels that are all
NaN (not-a-number):
|
Fixed. | 50-022 | F | 5.0_10, 21 Nov 2008 |
| B | When writing ROI statistics, header for statistics showing the mean statistics for all ROIs contained "NaN" (not-a-number) for the mean and standard deviation when all ROIs have no area (such as Markers). | Fixed. If all ROIs have no area, then the mean and standard deviation shown are the mean and standard deviation of the intensities under the points. | 50-021 | F | 5.0_10, 21 Nov 2008 |
| b | Incorrect status message shown in main display frame when ROI statistics and intensities have been written. | Fixed. | 50-020 | F | 5.0_10, 21 Nov 2008 |
| B | Writing of ROI statistics failed when wring statistics for floating-point images. | Fixed. Error in calculation of intensity Histogram bin width for floating point images. | 50-019 | F | 5.0_10, 21 Nov 2008 |
| b | ROI contouring could be initiated by pressing the 'c' key on the keyboard, even if the displayed image was not slice-selected. | Fixed. An image must be slice-selected before contouring can be initiated. | 50-018 | F | 5.0_10, 21 Nov 2008 |
| e | Enhancement to work around problem caused by a bug in a software package Applicare RadWorks Version 6.0 on GE Medical Systems scanners, which writes multiple null values after valid DICOM data. Jim interpreted as tags <0x0000, 0x0000> and therefore out-of-order element tags. The work-around attempts to detect these out-of-order Tags which follow after the pixel data, and terminate reading normally. Please report any further problems with using Jim to read images from GE CD-ROMs, or if this work around has created further problems with reading/converting DICOM images. | - | - | - | 5.0_9, 13 Nov 2008 |
| b | Jim failed to read illegal NIFTI-1 one-file compressed images produced by FSL 3.3, where the image header indicates that there are no NIFTI extensions, but where the image data did not follow immediately after the header. This problem is related to 50-015. Please report any further problems reading NIFTI-1 images produced by FSL. | Fixed. | 50-017 | F | 5.0_9, 13 Nov 2008 |
| b | Rotation of rotatable ROIs does not work for Jim running under Microsoft Windows. When attempting to rotate an ROI, the ROI flickers and the rotation is unsuccessful with the status message "ROI rotation cancelled". | Fixed. | 50-016 | F | 5.0_8, 3 Nov 2008 |
| b | Jim incorrectly read corrupt NIFTI one-file images produced by FSL 3.3 with missing extension data. Corrupt files can be identified by the fact that they are 4 bytes too short and the image data starts at byte 348 rather than 352. Jim has been enhanced to read corrupted files like these transparently. | Fixed. | 50-015 | F | 5.0_7, 24 Oct 2008 |
| e | Enhancement to Interoperability Preferences in Jim's main Preferences dialog. A new preference has been supplied to allow newly-created Analyze images always to have a non-flipped orientation set in the header, so that images produced by Jim can be viewed correctly in external programs such as SPM, MRIcro and Image/J. These programs may not correctly recognise the Analyze flipped orientations. | - | - | - | 5.0_7, 24 Oct 2008 |
| e |
New keyboard accelerators have been added for paging through slices:
|
- | - | - | 5.0_7, 24 Oct 2008 |
| e | Users have reported problems with the movie image in the Splash Screen slowing down startup when Jim is being run across a network. The splash screen has been made static. | - | - | - | 5.0_7, 24 Oct 2008 |
| e | Text ROIs can now have individually-coloured text. Previously, all Text ROIs had to have the same colour of text, which was also the same colour as annotation text. Now, all annotation has the same colour text (set in the user preferences), but Text ROIs can be individually coloured, just like other ROIs. | - | - | - | 5.0_6, 29 Sep 2008 |
| e |
Improved saving of ROIs to disk file:
|
- | - | - | 5.0_6, 29 Sep 2008 |
| b | When typing in the text for a Text ROI, the computer beeps on every character typed. | Fixed. | 50-014 | F | 5.0_6, 29 Sep 2008 |
| b | With linked Master/Slave displays, adding an ROI to the Master could result in the selected slice being changed in the Slave display, if the Master and Slave were not showing the same slice. | Fixed. | 50-013 | F | 5.0_6, 29 Sep 2008 |
| e | Enhancement to work around problem caused by a bug in a software package "Centricity Digital Hardcopy v1.0" used to create DICOM archive CD-ROMs on GE Medical Systems scanners. This software generates invalid DICOM files with Element Tags out-of-order. The work-around attempts to detect and ignore these out-of-order Tags. Please report any further problems with using Jim to read images from GE CD-ROMs, or if this work around has created further problems with reading/converting DICOM images. | - | - | - | 5.0_5, 02 Sep 2008 |
| b | Saving 3-D geometry to VRML file popped up a File Chooser for opening a VRML file, rather than saving a VRML file. On a Macintosh platform, this did not allow the user to create a new VRML file. | Fixed. | 50-012 | F | 5.0_5, 02 Sep 2008 |
| b | Drag 'n' drop of images and files does not work on Macintosh platform. | Fixed. Please report any further problems with drag 'n' drop, or with the Macintosh platform in general. | 50-011 | F | 5.0_5, 02 Sep 2008 |
| b |
On Macintosh platform, attempting to perform various operations in
Jim's 3-D Viewer, such as changing the material colour,
transparency etc. for orthogonal planes, iso-surfaces, displayed ROIs
etc. results in a java stack trace such as the one below:
|
Fixed. | 50-010 | F | 5.0_5, 02 Sep 2008 |
| b |
In Jim's 3-D Viewer, if a displayed iso-surface has a cutout,
then attempting to save the geometry to a VRML file results in a java
stack trace:
|
Fixed. | 50-009 | F | 5.0_5, 02 Sep 2008 |
| b | Analyze/NIFTI images with files names that have an uppercase extension (such as .HDR or .NII) cannot be opened. | Fixed. | 50-008 | F | 5.0_5, 02 Sep 2008 |
| e | Enhancement to Interoperability Preferences in Jim's main Preferences dialog. A new preference has been supplied to allow Analyze images in which the orientation set in the header is one of the "flipped" orientations, to be treated by Jim as though the orientation were the equivalent non-flipped orientation. | - | - | - | 5.0_4, 1 Aug 2008 |
| b | During contour ROI creation in "Preview contours" mode, sometimes a "zombie" ContourROI is left on the screen which is seen on every slice, and which will not go away until new ContourROI generation is initiated. This bug is hard to reproduce, and seems to be associated with a particular combination of key-presses. This bug does not affect quantitative ROI values (e.g. calculated volumes). | Code re-factored to try to avoid this problem; key-presses which could trigger this have been prevented and instead a warning beep and warning message are produced. Please report any further problems with this bug. | 50-007 | C | 5.0_4, 1 Aug 2008 |
| b | Pixel sizes for Bruker ParaVision™ images are too small by a factor of 10. | Fixed. Field of view in "reco" file is in cm, not mm as thought. | 50-006 | F | 5.0_3, 24 Jun 2008 |
| b | Image Converter failed with DICOM images containing Sequence items with VR set to UNKNOWN VR in the DICOM header. | Fixed. | 50-005 | F | 5.0_3, 24 Jun 2008 |
| b | Image Convertor does not correctly convert R,G,B DICOM images when "Save to disk" is selected. | Fixed. Incorrect coercion of data types was taking place. | 50-004 | F | 5.0_3, 24 Jun 2008 |
| B |
Masking of images using Masker tool did not perform correctly when image
slice contained no ROIs, or contained ROIs that enclose no area (Markers,
LineROIs and CurvedLineROIs). No masking took place.
Also, masking from within ROI toolkit did not function correctly when masking using a Markers, LineROIs and CurvedLineROIs (masking has no effect). |
Fixed. | 50-003 | F | 5.0_3, 24 Jun 2008 |
| b | Java stack trace shown when using image converter, selecting images or
DICOMDIR file from a CD-ROM, and the CD-ROM is ejected and a new one
loaded. Stack trace looks like:
|
Fixed. Check for invalid folder due to removable media has been implemented. | 50-002 | F | 5.0_3, 24 Jun 2008 |
| b | Image Converter failed with DICOM images from a Philips scanner will empty sequence elements. | Fixed. Check for empty sequences performed. | 50-001 | F | 5.0_2, 13 Jun 2008 |
| - | New release. | - | - | - | 5.0_1, 05 Jun 2008 |