Most Common Issues as Reported at SIIM 2010

E-mail Print PDF


Share

Skip Kennedy, Kaiser Permanente talked about the top PACS problems. An important issue is what to do if the system goes down. One of the things people focus on is to get a better reliable system, in other words, how to get more "9’s in your uptime (99,999%?). However, each "9" relates to a substantial amount of investment and money. It might be better and more cost effective to come up with alternate workflows, a good test system, work- around, because a system will go down one time or another.

Performance is an issue as well; as Skip mentioned, most people never had a call about the system being too fast, but many people are complaining about it being too slow. A stopwatch is a tool that every pacs administrator needs, but should be the last resort as real-time data about performance is a must and often not readily available. Real time performance dashboards are a must, however require significant IT investment. These type of investments can be made by major institutions, however, open source tools, and built-in tools are needed. A major issue is the disparity of the systems and data, e.g. if one wants to know the steps involved with a simple exam, one might need to do data mining of the HIS, to find out when it was ordered, the RIS, when it was scheduled and completed, the modality, when it started, the PACS when it was read and the voice recognition system when it was signed off, reported and faxed or securely emailed to the physician. This leaves open the time that it was actually be read by that physician.

David Clunie from Radpharm also gave some examples in case the PACS system does not perform as you might expect. One of the examples that he gave was with regard to the display of annotations. For example, a drawing on the image at thon modality or workstation might not be visible in the  CD viewer, in the web viewer, or another PACS or EMR. This is despite the fact that everyone is DICOM and/or IHE compliant. Overlays can be sent a) in the pixel data, i.e. "burned in", b) in the image header, or c) as a Presentation State, all of these solutions being totally DICOM "legal" but if one system uses method a) and the receiver uses method b), there is no true interoperability. To diagnose this problem, one could send the images with known annotations to simulator, run validation tools, send test images to the receiver system using the sets of test images used by IHE and tweak and/or modify sending systems.

Share
These signals are relayed buying clomid online safe which then is by a number of such as medial preoptic and paraventricular nulcei.
javporn.cc