Data storage, medical recordsTackling the PACS data explosionThe high data storage requirements of digital imaging presents significant IT management problems for hospitals. Charles Mallio of BridgeHead Software discusses the issues of cost, energy use, backup, archiving and compliance. October 2008
Picture archiving and communication systems (PACS), are now in use in every NHS trust in England [1, 2] as a result of the NHS National Programme for IT. PACS deliver many benefits for medical imaging, from improved image access and speedier diagnosis to cost savings from not having to store film. But hospitals face significant IT and administrative challenges related to storing and managing the huge volume of digital images that are generated. Intelligent PACS data archiving systems are emerging to tackle these problems. Supplied both by the PACS manufacturers themselves and by independent software companies, the archiving systems can help contain rising storage management costs as well as ensuring PACS data is effectively protected and secured in line with strict regulations, while remaining easily accessible over the long term. Rapid growth in image data The root cause of PACS storage management problems is the rapid growth in image data, which partly results from the ever widening application of PACS systems. When initially introduced, PACS technologies were mostly focused on radiology. But they are now increasingly being applied to CT, MRI, sonograms and DSA. In the UK, the government's new five-year cancer strategy promises £100m for digitising breast screening by 2010. The widespread escalation in image data means the cost savings of digitising could quickly be consumed by IT storage and management costs. Improvements in the quality of scanning technology are also contributing to data growth. In the past, typically around 20 to 30 images would be retained per CT scan. But today’s spiral CT produces 100 to 300 images and with a multilayer CT, this number can jump to 1600 images per patient. An average hospital with CT/MRI can produce about 500GB of data in a year, which needs to be stored for many years to come. And larger institutions can generate up to 5 Terabytes (5,000 Gigabytes) per year. Cost worries In the past, PACS data storage costs were not something that hospital management was overly concerned by. With prices for primary data storage declining, they assumed you could just buy more storage hardware and not worry about the problem. But physical storage costs — especially relating to spinning disk based primary storage — are compounded by significant software and management costs. This includes the systems and processes required to maintain the growing volume of confidential data in line with compliance regulations governing data security and protection. A major factor is ensuring that PACS data is regularly backed up, often to multiple copies, so that it can be quickly restored in an emergency. Backups have to be scheduled out of hours, often over night, to avoid disrupting operations. And as data volumes expand, backups become more difficult to schedule and are increasingly prone to errors. The whole process can eat up huge chunks of management time and resources. Rising energy bills The other issue with using disk-based primary storage is the high and rising cost of energy. These systems have to be kept powered on and require cooling round-the-clock, so electricity bills are not insignificant. And there is increasing pressure on public sector organisations to be more energy conscious in order to reduce carbon footprints. Intelligent PACS archiving systems address these storage issues by migrating PACS image studies that are no longer active to less costly storage media. The historical data can be stored on off-line media such as optical disk and magnetic tape, for which the operational cost is negligible compared to the costs of using primary data storage. By freeing up the primary store the management burden is greatly reduced. Removing infrequently accessed data This kind of archiving strategy can have a significant impact. In most file systems up to 80% of data held on primary storage consists of ‘static’ data that is no longer being frequently accessed or modified. A similar ratio is thought to apply in the PACS world. After all, when patients have undergone successful treatment, the corresponding PACS studies are very infrequently accessed. Most intelligent archiving systems are based around a policy-based software engine which automatically identifies historical, inactive PACS images in line with specific rules defined by users. The images are removed from primary storage and placed on the appropriate secondary, off-line media, which is managed via robots and automated for 'near line' access. Users who need to view images from the archive will notice no discernable difference. In fact, some archiving systems allow the data placed in the archive to be viewed through the hospitals’ PACS software in the usual manner. Clinicians need never even realise that data is stored within the archive. All they know is they can view the data they need when they need it. Streamlined backups By reducing the overall volume of data on the primary store, an intelligent archiving strategy such as this can streamline backup and replication processes, cutting the resource and management overhead they require. The most effective archiving systems can be configured to create multiple archive copies of the historical image data in multiple locations, which means the archive will not require frequent backups. The hospital management need only concern itself with backing up the smaller proportion of live PACS data which remains on primary storage. One point of confusion surrounds why it is necessary to have both backup and archiving. The reason is simple. The two are designed to solve different IT problems. Backup is intended to provide the ability to fully recover all or major parts of an IT operation in the event of a significant problem or disaster. Archiving, on the other hand, provides value via a more granular, file level approach — offering easy user access to individual image files. Files placed in an archive are indexed and searchable by content and the metadata relating to the PACS images. Some advanced PACS archiving systems now index using the patient-specific header information contained within each image according to the DICOM (Digital Imaging and Communications in Medicine) standard used by most PACS vendors. Hence the archive becomes a searchable store, in which it is easy to locate specific PACS studies. By contrast trying to locate single files from a backup is fraught with difficulties. Designed to handle compliance Moreover, many archiving systems are specifically designed to manage data according to compliance regulations. This includes securing information through data encryption and using WORM media and digital signatures to control access to PACS data and verifying that it has not been altered or corrupted. Specific policies can be defined to automatically delete data at end of life, including crypto shredding techniques to ensure there is no way of deleted data falling into the wrong hands. Today, many leading manufacturers of PACS systems are starting to offer proprietary solutions to archive PACS data. But these aren’t necessarily the most efficient option because they ‘lock you in’ to one manufacturer’s PACS offerings. Independent archiving suppliers offer solutions which work with any number of PACS suppliers and with whatever storage platform. In fact, intelligent archiving products that make use of the DICOM standard actually help to improve overall accessibility of PACS data. They allow clinicians from different departments and hospitals — which may rely on disparate proprietary PACS systems — to view and share images through an archive, regardless of the underlying PACS manufacturer. This is becoming increasingly useful. Especially as the NHS strives to enable images to be shared 'seamlessly' across clinical networks or exchanged with other centres when patients travel. While the NHS National Programme for IT has succeeded in getting PACS systems into NHS trusts, one of the next steps must be to ensure the data generated is managed effectively and kept accessible and secure over the long term. Intelligent archiving systems have a significant role to play in making sure this happens without ‘breaking the bank’. Charles Mallio, Director of Business Development — Healthcare Sector, BridgeHead Software. References 1. Health Service Journal. 11 February 2008 www.hsj.co.uk/goodmanagement/bestpractice/2008/01/pacs.htm 2. First phase of PACS rollout to all NHS trusts in England completed.bjhc&im news, 14 Jan 2008. www.bjhcim.co.uk/news/2008/n801013.htm |
Please allow
scripts in your browser so that Google ads will show — the ads
are safe and give information on useful IT products.
|
||
|
|
|||