In our previous discussions about healthcare IT on this blog, we've covered an overview of typical architecture that is found hospitals or clinics, and how to monitor it. More recently, we started looking at individual components (like the Integration Engine). Now, in this article, we take a look at another vital element: the Picture Archive and Communication System, or PACS.
Essentially, a PACS is an image sharing platform. Modalities (healthcare IT jargon for "imaging devices") take all kinds of pictures of patients: X-rays, ultrasounds, CT images, and so on. These images are saved to a central repository, where they can then be accessed or transferred by other workstations or devices. This central repository is the PACS. A PACS also archives older images that need to be kept on record.
Because of how central images are to the diagnosis and treatment of all kinds of conditions and injuries, the PACS is an important element of the IT infrastructure. Here's a paragraph from our Healthcare IT Monitoring Primer about why it's so important:
In a typical hospital setup, the PACS connects to the modalities, the Radiology Information System, and the integration engine. It forms a central part of many radiology workflows, which subsequently means that problems with the PACS can have a big impact on the general functioning of the hospital.
Basically: if you can't store or access images, many workflows in the hospital will be interrupted. To ensure that it remains up and running, it needs to be constantly monitored for potential problems, and to alert IT professionals when something fails. Here are four aspects of the PACS that can (and should) be monitored using network monitoring software.
Because it is primarily a storage system, it requires disk space. This should be monitored and thresholds set to alert IT professionals when storage space is getting low. Also, the hardware that the PACS runs on must be monitored should be monitored for signs of hardware failure (like overheating, no more available RAM, and so on), as well as to check that they are up.
The PACS is constantly storing and retrieving images to and from the storage system. This includes tasks like pre-fetch, short-term transfer, long-term transfer, and more. Low latency is required for any action between the PACS and the storage system, because higher latency will cause the entire system to become sluggish. It's recommended to monitor this latency and set thresholds for when any of the PACS tasks take too long.
Many PACS provide some kind of API for accessing information about the health and status of the components, and almost all of them create log files. A typical PACS API will give you statistics about the current application performance and metrics like number of DICOM requests received, number of errors, and the status of internal processing queries. Depending on the API in question, you can use a network monitoring system to retrieve these metrics (for example, using REST queries if the API provides a RESTful Interface) and create alerts when values move out of the expected ranges.
Meanwhile, log files provide indications of problems like failed authentication attempts or internal PACS failures. It is highly recommendable to ensure that a monitoring system is regularly checking the log files for specific events that might indicate problems.
Because the PACS is so central to the architecture of a typical hospital's IT infrastructure, it has interfaces to multiple systems and devices, like the Radiology Information System, imaging devices, and more. So it makes sense that these interfaces should be closely monitored.
There are two protocols that are commonly used for communication between healthcare systems, and they can be used to monitor the status of the interfaces:
Of course, there is much more to a healthcare system than just the PACS. To learn more about the other elements, and also how they should be monitored to ensure that your healthcare IT systems remain...erm, healthy...download the primer below. And if you have any questions or thoughts, let us know with a comment below.