PACS data migration projects are comprised of the following elements:
- Inventory – An inventory of the source system is taken, either from the source system’s database or through its DICOM Query interface. The inventory also extracts current values for Patient and Study attributes from the legacy system. Also, if the project includes data cleansing, an inventory of the imaging orders is extracted from the HIS/RIS.
- HIS/RIS Data Reconciliation – This step is the matching part of data cleansing. Matching must be performed before a study is moved. Laitek migration tools allow migration of matched studies while other studies are still awaiting matching operations.
- Source PACS extraction – Image data is extracted from the legacy archive, by reading off-line tapes or optical disks, by FTP from the legacy archive’s RAID, or by DICOM query from the source (legacy) archive.
- Processing – Images read from the source PACS may require processing before being sent to the new PACS. Processing is required when image data is read directly from the source PACS storage.
- Sending –The processed data is sent to the new PACS. This is normally done in one or more DICOM Storage streams. The sending process may also include conversion to or from compressed Transfer Syntaxes, as required by the destination PACS. Optionally, the sending process may be customized for the specific destination archive to achieve higher throughput.
- Verification – Data is queried from the destination PACS to verify that it has been received correctly. Verification may include direct image-to-image comparison between source and destination PACS (usually spot-checked), and Study- or Series-level comparison of image counts between the destination system and the source inventory.
With those elements in mind, let’s next consider the planning of your migration and go-live schedules.