Xero Display Area Architecture

The Xero display area follows the general patterns from the Xero Web Page Design, but of course it has specific pages that are different from the other areas. There are currently 4 primary frames that are displayed. The Study frame contains information about the general studies that belong to the given patient, and allows control over which study is displayed, and whether the study is displayed as a report, key objects or full-study images. The Series frame contains information about each series, and allows some amount of control over how the series are displayed. The series that are displayed are either the study series, or the key objects. Finally, the image frame displays one or more images, and has controls over the image level display.

Study Frame

The study frame is a relatively simple page. On the left hand side it has a study selector, while on the right hand side it has a set of study commands, including series, key-objects and report to cause each of those items to display. This frame just takes the cycle list xml as the indication of how to display, and uses a set of sub-frames to display the series tray and image/report area.

Series Frame

The series frame is somewhat more complex. The upper portion contains series level commands, potentially including items such as screen layout for the image area, as well as paging options for the series tray. Below that area all of the series are listed, each with an example image.

The example image is somewhat interesting in terms of design. A simple series level query is required in order to fetch all the series level meta-data, but this doesn't retrieve any example image meta-data, and there isn't a web-call to retrieve "some image in this series". As well, not all series should be shown, and some series need to be broken down into separate series. Overall, this is handled by a number of different filters, and can cause some studies to require several queries to retrieve enough meta-data in order to display. The filters are:

• series xml filter. This does an IMAGE level query on the study, by Study Instance UID, with the additional requirement of only returning the first dicom object in each series. For many studies, this will be sufficient and will return the appropriate instance study.
• MR Echo Split (Todo). is a filter that splits MR series by echo time. Often MR series will have first, second and third echo images all incorporated into a single series, but logically these represent the same image location but different types of views. These are usually split into separate series. This filter may take some amount of time to run as it requires a fairly expensive complete image query to perform.
• MG Split and Regroup (Todo). Some manufacturers group all or some MG studies into a single series. However, the viewer usually wants these grouped by view position, and with retakes stacked. This requires re-querying for all the image information and then regrouping it. This isn't much extra effort to run as MG studies do not contain many images. For mixed-modality studies, this query will perform a complete query for the MG series only, and use the regular processing for non-MG series.
• Manual split (Todo). This would be a possible filter to support manually splitting studies.
• Merge Series. This would be a possible filter to merge two or more series into one.
• Merge Images. This would be a filter to merge individual images into a single series, and would typically be done manually.

Key Objects Series Frame

The key objects series frame is just a customization of the regular series frame that splits series up into key objects rather than showing the original series. There are some custom additions to the commands available when key objects are being displayed. Those are the ability to select which key object is displayed, and the ability to switch to the regular series tray.

Image Frame

The image frame is quite a complex frame with multiple layers for different purposes. The layers are: Full-frame, used to contain the full frame information, for things that aren't aligned with the image. Demographics contains the demographics information for the patient, study, series and image. The Image aligned layer contains the image, the markup, the overlay and the shutter layers. These are handled separately as can often be downloaded separately to improve performance re-use. There may also be a new-markup layer for modifications to existing markup. Each of these will have a separate XSLT document, and a command to modify these can modify/extend all of the XSLT documents, or just one or several of these.

The image xml contains the images for just the selected series. This can be modified by using a custom SeriesInstanceUID in a format the filter understands to perform operations such as splitting a series etc. The filters for the series level need to be matched by image level instances that understand the appropriate operations. For example, an echo-split might have the form SeriesInstanceUID=EchoSplit(1.2.3.4,2) where the 1.2.3.4 is the original series UID, while 2 is the echo number.

In addition to the image xml, at some point gsps or markup XML will be required. This will need to list the annotations, overlays etc to use. It isn't clear yet whether to incorporate this into the image xml directly, or to look this up on each image lookup. Likely some amount of performance measurement will be required in order to understand this better.

Report Frame

The report frame is a fairly simple frame, just using existing services to retrieve the report as PDF, HTML, RTF or other formats that the browser can display.