DICOM compatibility
DICOM Standard
The DICOM file format is commonly used by medical CT and MRI scanners of the kind in widespread use in hospitals and other medical centres. However a growing number of micro-CT and nano-CT scanners are offering DICOM as one of the formats that scan data sets can be saved as.
Disect Pro can accept scanner data sets saved in DICOM format. However it should be noted that the DICOM standard, though comprehensive and well documented, can be interpreted and implemented by scanner manufacturers in slightly different ways. The Disect Pro development team have verified compatibility with a number of different scanners, however should you experience difficulties in viewing your DICOM data sets in Disect Pro please contact us and supply an example of the affected data so that we can investigate.
Loading Data
File handling
Disect Pro is designed to work with the following file formats :-
DICOM (12bit and 16bit)
BMP (8bit and 16bit)
TIFF (8bit and 16bit)
JPEG (8bit and 16bit)
JPEG2000 (8bit and 16bit)
That range extends with each new release of the software. The bit resolution that can be accommodated in each format is shown in brackets.
1. Opening a DICOM file
The opening procedure for a DICOM file might depend on the type of scanner used – a human size medical scanner, or a micro-CT or nano-CT scanner.
In the case of a human size medical scanner the DICOM data is commonly stored using the documented DICOM Directory structure. Data may be loaded from the local hard drive, a networked drive, or from CD-ROM or DVD. Disect Pro expects to find a file named ‘DICOMDIR’ in order to load the data automatically from CD/DVD. If you do not have a DICOMDIR file on the CD/DVD, i.e. you have just a number of DICOM files forming a data set, Disect Pro will automatically attempt to scan the CD/DVD to locate any DICOM data sets.
DICOM data from a micro-CT or nano-CT scanner may not contain a ‘DICOMDIR’ file in which case the second procedure described above should be used.
Loading DICOM data from CD or DVD

The CD/DVD may optionally have a DICOMDIR file in its root folder. This is the format most commonly produced when burning a CD direct from a medical scanner workstation. Insert the CD/DVD into the CD-ROM or DVD drive, then select File / Open CD from the main menu, or click the CD icon on the toolbar. Disect Pro will scan all of your CD/DVD drives looking for DICOMDIR. If that doesn’t exist Disect Pro will scan the CD/DVD for DICOM data. If it finds some data, continue from ‘Selecting the DICOM data set’ below.
Note: this method should only be used to open data in DICOM format directly from a CD/DVD. To open all other formats directly from a CD/DVD please refer to the instructions below in section 2 for the other file formats.
Scanning a folder

If you have a number of DICOM data files in a folder on your hard disk, Disect Pro can scan this folder (and all of its sub-folders) and collect details of all DICOM data sets it finds. After launching Disect Pro , select File / Scan Folder from the main menu, or click the relevant icon on the toolbar. Disect Pro will prompt you for the location of the folder and will scan this for for DICOM data. If it finds some data, continue from ‘Selecting the DICOM data set’ below.
Loading DICOM data directly from disk

To load data directly from disk, either select File / Open file(s) from the main menu, or click the folder icon on the toolbar. If you do have a DICOMDIR for the data you wish to load, navigate to that file using the FileOpen dialog as shown below. If you do not have a DICOMDIR, navigate to the folder containing your DICOM images and select any one of the files from the data set you wish to load. Disect Pro will automatically locate all remaining files for that data set in the same folder and will load them in the correct order. If you have previously saved your data in Disect Pro’s native format (*.dsv), simply choose the file you wish to load and click Open.
Example of the FileOpen dialog:

Selecting the DICOM data set
A DICOM Directory (DICOMDIR) or file set may contain data for more than one scan. Each data set is identified with some basic information – patient name, patient id, procedure description, date, time, etc. When loading data from DICOM files, you will be presented with an additional screen, which allows you to choose the data you require. This will list the patient name and id along with the study name, series and other data as shown below:

Note that, in the above display, the patient name and id are anonymised. You can display the underlying name and id by switching off the anonymise option from the View Menu.
Select the data set you wish to load by clicking on the entry in the list, then click the Load button to proceed. Double-clicking an entry in the list has the same effect as clicking the Load button.
Managing the size of the DICOM data set
Disect Pro is able to downsize a large DICOM data set during image reconstruction, just prior to being displayed. That might reduce image quality so it is recommended that the data set be reduce at the import stage. When the particular scan has been selected in the above dialog box, the user is given the option to select either the whole data set (i.e., all slices in the stack) or just a subset, as shown below:

In the above example the user has selected slices 121 thru 437 from a larger set. The resulting number of selected slices is displayed (317) together with an approximation of the data size that Disect Pro needs to support in order to display the selected slices in full resolution. The size in this example (165MB) is relatively small and so will be displayed in full resolution on even a modest PC.
2. Opening a BMP, TIFF, JPEG, JPEG2000 file
The opening procedure for a BMP, TIFF, JPEG or JPEG2000 file begins by selecting ‘import greyscale images’ in the FILE menu. In the window that appears (see below) click on ‘add files’ and navigate to the folder containing the data set you wish to open. Select the slices in the data set you wish to open and click on ‘open’. These slices will be listed in the vertical column to the left of the window. Typically you need not change the HU (Hounsfield Unit) bit depth range or the pixel dimensions or slice spacing that appear in the window, but you can do so at this point by entering the desired number in the boxes. You also have the option at this point to reverse the numbering sequence of the slices as they are loaded into Disect Pro. That can be useful since there is no numbering convention and some scanners place slice number 0000 at the base of the scanned object and count upwards, while other scanners place slice number 0000 at the top of the object. When you have made all your selections click on OK.

In the next window (see below) the quantity of slices you have selected is displayed, together with the dimension of each slice in pixels. At this point you have the option to view the slices in Disect Pro at their default size or to downsize them by entering the desired number in each box. For example, a data set from the scanner comprising slices 1024×1024 pixels in size can be downsized to 512×512 by entering 512 in each box. The ability to downsize at this point in the file opening procedure has been incorporated into Disect Pro to give the user to means to reduce the overall size of the date set within Disect Pro. That in turn speeds up the rendering process and the smoothness of the viewing experience. Applying too much downsizing will however reduce image quality. The optimum choice can be determined by trial. With your selection made click on OK.

With all the selections now made the data is imported into Disect Pro in 3 phases: loading the image slices, building the cache, building the image. The progress of each phase is displayed by a horizontal bar. Upon completion of the 3rd phase the data will be displayed in Disect Pro.
Note: Disect Pro is designed to display images in 3D that are reconstructed from multiple 2D slices imported as BMP, TIFF, JPEG or JPEG2000 files. However Disect Pro can also display a single 2D image, such as an X-ray or a photograph imported in one of these formats.
Note: Grey scale as well as colour 2D images can be imported however colour information is stripped off within Disect Pro and all displayed views are in grey scale. To maximise image quality it is recommended that colour 2D images are first converted to grey scale before importing to Disect Pro.
3. Opening a file previously saved in Disect Pro’s *.dsv format
At any point when viewing a scanner file in Disect Pro it can be saved using Disect Pro’s native format (*.dsv) which incorporates a proprietary lossless compression algorithm.

To open a *.dsv file in Disect Pro either select ‘open DICOM or dsv file’ in the file menu, or click on the folder icon on the toolbar.
Subsampling
What is subsampling?
Subsampling enables very large data sets to be imported while staying within the upper memory limits set primarily by the available graphics memory. It achieves that by loading into Disect Pro only 1 slice in every N slices that appear in the import window (see below) where N is determined by the user.

Begin by importing all the slices in the data set, or those that span the desired segment of the scanned object. These slices appear in the import window (above). Next select “Subsampling” to open the window below.

A default value for N, the “subsampling ratio”, is pre-selected which ensures that the resulting number of slices in the reduced set is approx 100. The user can increase or decrease that figure by moving the horizontal pointer. Increasing the number of resulting slices improves the displayed image quality but reduces the area of the scanned object that can be accommodated within the graphics memory. Reducing the number of resulting slices reduces the displayed image quality but, depending on file size, may allow the whole of the scanned object to be viewed.
When opening a new data set for the first time the user is advised to use the default subsampling ratio (and also re-sizing if needed) in order to view the whole scanned object. If there is sufficient headroom within the available graphics memory, the subsampling ratio can be reduced on subsequent passes to improve displayed image quality.
Downsampling
What is downsampling?
Disect Pro conducts the majority of image processing and surface rendering functions within the dedicated graphics engine in the host PC. That is done to take advantage of the superior power of the graphics engine. It is therefore vital that the data set for the scanner file to be loaded can all be accommodated within the memory capacity of the graphics card. In circumstances when the size of the data set exceeds the graphics memory, it is necessary to downsample the data set. In essence, downsampling reduces the number of pixels in each 2D slice in the displayed image.
Automatic downsampling
Disect Pro automatically detects the size of the file being loaded and compares that with the available graphics memory. It then decides on how much, if any, downsampling to apply: 2x, 4x, 8x or 16x.
NOTE: Downsampling is only applied to the 3D volume view, the 3D MIP view and the 3D cut view. The 3D MPR view and all the 2D views are always displayed with no downsampling applied, i.e., they are displayed at full resolution. That is because there is substantially more image data in the 3D volume, 3D MIP and 3D cut views compared with the others which must be accommodated in the graphics memory. The significantly less image data in the 3D MPR view and 2D views means that they can be much more readily accommodated.
Whenever downsampling has been applied to a file being loaded, the degree of downsampling appears above the 3D window. If downsampling has not been necessary ‘Full resolution’ is displayed above the 3D window.
Maximising image quality
Downsampling inevitably degrades the displayed quality of the 3D volume, 3D MIP and 3D cut views. If only 2x downsampling is applied the reduction in image quality is minor, and in some circumstances may not be detectable, depending on the content of the image on display. At greater degrees of downsampling the reduction in image quality becomes more noticeable, and takes the form of a blurring of fine details. To minimise the need to downsample, and therefore maximise image quality under most circumstances, it is recommended the host PC on which Disect Pro is being operated is equipped with a dedicated graphics card with the maximum available/affordable memory.
Saving Data
At any point when viewing a scanner file in Disect Pro it can be saved in Disect Pro’s native format (*.dsv) which incorporates lossless compression. A typical compression ratio might be 2x for images that contain a good amount of detail and few ’empty’ regions, increasing to perhaps 6x or more for images with less detail and more empty regions.
Saving a file within Disect Pro conveniently produces a single *.dsv file regardless of the number of slices that have been imported. This greatly simplifies file handling and data storage requirements.
To save your data select File / Save As from the File menu and enter a filename in the dialog box.