We have migrated from one PACS to another using a mix two methods:
- copying images to NAS drives and driving them to the new data centre, and the new supplier querying the old suppliers DB, then re-creating the DICOM files and pushing them using DICOM into the new VNA
- DICOM QR from new PACS to old PACS
We are now live on the new PACS and looking at end Jan before all of our legacy data is pushed into the new PACS.
What might be the down-sides of terminating the old PACS support contract in September and just relying on the the copies of data we have pulled from it?
Difficult to say. In 2013, moving from supplier G to supplier F, for the data migration, we used a copy of our data that supplier G kindly provided in an expensive box. Supplier G was not further involved in the data transfer.
For our interest, how much data are you moving and how much time have you reserved to complete the data migration? I’m trying to get some idea how much time these processes take. We know it depends on transfer method, data cleanliness, supplier etc, but as a starter, from a technical perspective, how much data can you move per day? The rest are variables on top. Need to have at least one variable that is calculable.
In this case would you lose all access to the old PACS at the date of termination, or just the support?
Frozen archives used to be the way to do things - just prevent further ingest, prune off most users and let the old servers spin away until several years have passed without needing anything. It’s the way film libraries tended to be absorbed back into the wilderness
A consideration of migrating from the NAS alone is whether it supports the DICOM query retrieve and whether it has access to the patient demographics. Sometimes the NAS storage contains only a proprietary link or the accession number to link the images to the demographic data stored in the PACS. That’s one of the main differences between a NAS and a VNA - as to whether it can act independently as a DICOM store.
Another consideration with demographics is during live operation the PACS receives HL7 demographic updates from PAS (sometimes routed via RIS) - to ensure for example that if a former patient gets married and changes their name then their entire historic patient record is updated. Without updates the frozen archive will contain the old demographics. This may be an issue depending on the criteria you use for demographic matching when the data are ingested/merged into the new PACS record - as there will then be a mismatch on some fields.
Very good point about the ‘demographic drift’ over time!