Can I ask for advice about about what is meant by Desktop Integration. My understanding was that it involved a bidirectional link that allows the RIS application to invoke display of images in PACS on a common workstation, and allows RIS data to be invoked from PACS. We are currently in discussions with our supplier who is implying that the requirements for desktop integration are met if the RIS component can invoke the PACS images, but not the other way round. This would mean for example that, if examinations are put into folders for MDTs etc, that when it came to the MDT meeting the images could be displayed effortlessly from the PACS system but that the RIS data would not be invoked automatically, and therefore would not be available alongside them. I'd appreciate any comments.
That is the way our system works at Bromley i.e calling a Dictation or a Report for a given examination on the RIS automatically calls the relevant images on the PACS viewing software. I think a bi-directional link would be useful but, in my understanding, the desktop integration is primarily a reporting tool. What would be very useful would be a configuration facility whereby relevant previous examinations would also be displayed automatically; i.e. chest X-Ray series or follow up CT etc. However even with just the examination in question the reporting process is significantly faster, so much so that when reporting high volumes of plain films fatigue and eye strain become more of a factor.
Our PACS system (Rogan) does receive and store a copy of the verified report from the RIS which would, in part, address your query.
Philip, desktop RIS/PACS integration can be implemented in different ways. The national OBS clearly defines it as a bidirectional interface.
115.8.202 The workstations provided for desktop integration shall, according to user preference, be provided in one of the following configurations: · 2 display devices for viewing images and 1 colour display device for specialty information solution (in the case of radiology, commonly referred to as a RIS). · 1 display device for viewing images and 1 colour display device for specialty information solution (in the case of radiology, commonly referred to as a RIS). · 2 display devices for viewing images with the option of interacting with specialty information solution (in the case of radiology, commonly referred to as a RIS) on 1 of these display devices.
· Workstation display devices shall be controlled by a single keyboard and mouse. The PACS and specialty information solution (in the case of radiology, commonly referred to as a RIS) application shall be linked in such a way that examination information selected on the specialty information solution (in he case of radiology, commonly referred to as a RIS) display device shall cause relevant images to be shown on the PACS display devices according to the relevant default display protocols (DDPs) and selection of images/examinations on the PACS display devices shall cause the relevant specialty information solution (in the case of radiology, commonly referred to as a RIS) examination data to be displayed on the specialty information solution (in the case of radiology, commonly referred to as a RIS) display device. · There shall be a single log on to the services such that a user logging on to the PACS shall also be logged on to the specialty information solution (in the case of radiology, commonly referred to as a RIS) and vice versa, where it is appropriate for a user to be logged on to both systems. · All functions permitted to the user of the specialty information solution (in the case of radiology, commonly referred to as a RIS) and PACS shall be available (including digital dictation where installed) through the single workstation with desktop integration.
It all works so well in the armchair! Its real life which always causes the problems. There are very rare occasions when I need to access RIS information during a MDT. I wonder what Philip is envisaging. All authorised reports are copied from RIS to PACS and without this PACS would be impoverished, almost useless. The RIS/PACS link we are testing is for reporting mainly. Our RIS supplier is looking into developing worklists from RIS as we move to a paperless RIS environment. Our PACS does carry some RIS information in that, as soon as a patient has an appointment made for a follow up scan (for instance) the entry is mirrored in the PACS database with the date of the appointment. In MDTs this can be useful in informing the discussion about when a follow up is scheduled.
posted on Wednesday, April 13, 2005 - 06:28 pm
I am unsure that I would entirely agree with John over the proper repository for reports being the PACS unless the link allows for updating demographic info or updating revised reports.
Here in Southampton our view supported by the clinical governance lead. is that the only valid report was that which resided on the RIS and that the PACS via its Broker should pull that report to the PACS viewer when it needed to be viewed.
Our discussions with IT over the national programme would seem to imply that the report repository would be at the cluster data store and we have had great difficulty in elucidating how and if this can be updated if required. We see this as a high clinical risk.
John, are you saying that your PACS does not have worklists from a RIS? If so I would have thought that this is an important omission. It must make radiographic staff's lives very hard.
I am sure that the best future is a fused RIS/PACS system but we have always believed that the RIS should be master of the PACS and not vice versa.
Ivan, Demographics are updated ONLY in HIS, this is then cascaded to all the other clinical systems in real time. Anything less is second best. Having more than one place to update demographics would lead to duplicates and it would be worse than two or three X-Ray packets (at least you could sellotape those together!). Any report amendments are automatically mirrored in PACS once the amendments have been authorised by the reporter in RIS, so PACS is always up to date. Whether PACS pulls or pushes is probably immaterial. It is RIS which pushes reports to the broker and the broker to PACS in Norwich. I don't know quite how our long term store works but I think that any amendments are updated when an image is pulled from long term store, this should be how the cluster store works or similar.
Over worklists I was not quite lucid. It is reporting worklists I am speaking of here, not modality worklists. Without modality worklists PACS becomes unworkable I quite agree.
John, just to be picky I'd say ONLY being able to update demographics in HIS is second best. Ideally you would want to be able to update patient demographics in any application (HIS, RIS etc.), and for the changes to cascade to all other clinical systems.
In NPfIT, the patient demographic service (PDS) will provide the 'central' repository, and provide the mechanism for compliant systems to initiate this cascade. In practice it will be a mixture of cascade and update, as systems will be designed to check demographics with the PDS each time a record is accessed. As you say, this is how you think your long-term store works.
What I meant was that the PMI should be in HIS. In practice we can update from many of the departmental systems by means of a background link. So, if you attempt an update in RIS, for instance, it merely brings up the HIS demographic update page without the need to change systems.