UK Imaging Informatics Group
PACS and VNA workflow question - PACS... PreviousNext
UK Imaging Informatics Group > Questions & Answers > PACS Integration & Standards >
 Link to this message Huw Thomas  posted on Friday, June 06, 2014 - 01:51 am Edit Post Delete Post Print Post
Hi all, I have a knowledge blackspot when it comes to moving from a PACS only solution to a combined VNA + PACS solution.

I am hoping those of you who have real world experience here can clarify.

My understanding is that when moving to a VNA with PACS, the PACS archive becomes a short term archive, with the VNA acting as the long term archive. The short term PACS archive is kept up to date with autorouting and prefetching/prepushing from the VNA. I.e. all images (including priors) that need to be seen by the reporting clinician are available on the PACS, despite it being a short term archive.

I also understand that it is vital that both the VNA and PACS support IOCM, which allows the PACS and VNA to stay in sync. I.e. any changes made in the PACS are automatically applied to the VNA.

Now my blackspot comes with how the PACS archive remains a short term archive. What/who tells the PACS that data is no longer relevant and should be deleted. Does this come from the VNA? Those of you that have moved to a combined PACS/VNA, how do you handle this?

 Link to this message Nathan Gurgel  posted on Friday, June 06, 2014 - 03:31 am Edit Post Delete Post Print Post
Hello Huw,

You are correct in your understanding of the typical deployment model of PACS + VNA, with the PACS retaining a cache large enough to retain somewhere the range of 12 to 24 months and the permanent, long-term storage being manage by the VNA. You are also spot on with regard to both the VNA and the PACS supporting IOCM to minimize administrative effort in keeping the data synchronized.

The method of the PACS archiving to the VNA while allowing the short-term cache purging goes by many names depending on the PACS vendor. DICOM archiving, 3rd party archiving, store and remember, event based forwarding are terms used by various PACS vendors, but the mechanisms are similar.

What happens is that PACS will C-Store studies it receives to the VNA based on some event. It could just be receiving the study, but is often triggered by report finalization or other types of workflow events. Typically a DICOM storage commit is executed to ensure that all the images stored haven been received by the VNA and are indeed stored there. At that point, the PACS will update its database that the study is archived in VNA. That allows the study be eligible for purging from the PACS short term cache based on the established rules. Often some variant of FIFO with a last accessed rule.

For retrievals, the PACS then has a record in its database if the study is present in its short term cache, the VNA, or both. If the images are in the cache, it will retrieve them from that location or will issue an ad-hoc retrieval from the VNA if necessary by issuing a C-Move for the needed study.


Nathan Gurgel
Product Manager
TeraMedica, Inc.
 Link to this message Huw Thomas  posted on Friday, June 06, 2014 - 07:43 am Edit Post Delete Post Print Post
Hi Nathan,

Thanks for your detailed response, that is very helpful. I can see that the PACS requiring Storage Commitment SCU is very important here.

This brings me to another question, what is the typical workflow with a combined PACS/VNA solution:

1. Modality --> PACS --> VNA; or
2. Modality --> VNA --> PACS

I would have thought the latter would make more sense as that way when it does come time to switch to a new PACS you are not retesting dozens of DICOM connections, just the one to the new PACS. However, if the typical workflow is 'Modality --> VNA --> PACS', how does the PACS know the VNA has a copy of the data? Or is it simply assumed that any data that arrives from the VNA is marked in the PACS database as archived on the VNA, with any new data created on the PACS then exported back to the VNA using Storage Commitment.

Thanks again for your feedback.

 Link to this message Neelam Dugar  posted on Saturday, June 07, 2014 - 01:53 pm Edit Post Delete Post Print Post
Hi Huw and Nathan,

This is an important topic that you have hit upon.

This our architecture.

1. PACS is for primary display and reporting--image display must be within 3 sec. PACS holds 8 years of images. Hence we do not rely on prefetch for reporting.
2. VNA is a back-up archive and for Disasater Recovery too.
3. Modalities send to PACS for viewing and reporting
4. Once the exam achieves a verified status--i.e. reported, then the images will be pushed into VNA. We do not expect many changes to the study and images after the verified (report authorised status).
5. IOCM once available will help to communicate between Fujifilm PACS and Perceptive/Acuo VNA any changes in the studies.
6. VNA will also do the life-cycle management--based on the national guidance on data retention. It will communicate the planned date of deletion of study to PACS. This will be displayed on PACS for clinical users. If users wish to extend the date of deletion--for slow growing tumours, medico-legal etc etc, they can call the PACS office.

IOCM is key.

We have a Quarantine Archive of images from other hospitals come in via IEP-- which are kept for a period of 3 months. These do not go into VNA as they do not have our Hospital ID.
 Link to this message Huw Thomas  posted on Tuesday, June 10, 2014 - 01:07 am Edit Post Delete Post Print Post
Hi Neelam,

Thanks for your response and workflow description.

I would be very interested in seeing what others are doing here - I will start a new post on this...

Add Your Message Here
Username: Posting Information:
This is a private posting area. Only registered users may post messages here.
Options: Automatically activate URLs in message