Problem-oriented Prefetching for an Integrated Clinical Imaging Workstation

Abstract

Prefetching methods have traditionally been used to restore archived images from picture archiving and communication systems to diagnostic imaging workstations prior to anticipated need, facilitating timely comparison of historical studies and patient management. The authors describe a problem-oriented prefetching scheme, detailing 1) a mechanism supporting selection of patients for prefetching via characterizations of clinical problems, using multiple data sources (picture archiving and communication systems, hospital information systems, and radiology information systems), classifying patients into cohorts on the basis of their medical conditions (e.g., lung cancer); and 2) prefetching of multimedia data (imaging, laboratory, and medical reports) from clinical databases to enable the viewing of an integrated patient record. Preliminary evaluation of the prefetching algorithm using classic information retrieval measures showed that the system had high recall (100 percent), correctly identifying and retrieving data for all patients belonging to a target cohort, but low precision (50 percent). A key finding during testing was that the recall of the system was increased through the use of multiple data sources (compared with one data source), because of better patient descriptors. Medical problems and patient cohorts were more specifically defined by combining information from heterogeneous databases.

Hard disk online caches in picture archiving and communication systems archives: how big is beautiful?

“…Imaging of more than one body region within one study (for instance, tumour staging by means of both thoracic and abdominal CT scans, or whole body MRI) may constitute a labelling problem that results in prefetching difficulties [13,15,20]. Even when very sophisticated pre-fetching rules, using HIS data, are applied, precision varying between 50% and 100% is reported [16,19]. The underlying problem can be summarized thus: the smaller (and less expensive) that caches are designed, the more discerning a pre-fetching strategy must be in determining relevant prior examinations.…”

“…Retrieval of prior examinations causes high network traffic loads and additional burden on the LTA as well. Thus, pre-fetching usually applies sophisticated algorithms to select presumably relevant studies and block non-relevant prior studies at the same time [16,18,19]. In daily practice, inconsistent DICOM interpretation of different imaging modalities and inconsistent definitions of body regions (even within the same radiological institution) may lead to missed detection of relevant prior examinations.…”

“…Second, for institutions with a moderate number of examinations but usually high access to prior examinations, the best choice might be to archive on hard disks exclusively and to expand the disks or disk arrays on demand [3,15]. In contrast, for the common case of large PACS installations with typical access to relevant prior examinations, archiving of radiological images is usually structured in a hierarchical manner [4,5,16]. Recently generated images have been stored both temporarily in fast RAID caches and in slower but cheaper (costs per unit of storage space) long-term archives (LTAs) for the whole data-retention period [6,7].…”

“…In order to avoid time-consuming study loading times [7], almost all PACS use pre-fetching algorithms. The basic idea is to initiate study loading along with patient registration in the hospital information system (HIS), or at least in the radiological information system (RIS) [16,17]. Ideally, at the time of image interpretation of a (potential) current patient study, all relevant prior examinations should already be cached.…”

Internet Infrastructures and Health Care Systems: a Qualitative Comparative Analysis on Networks and Markets in the British National Health Service and Kaiser Permanente

“…Stable common ground shared by actors in the clinical decision-making system may be associated with asynchronous modes (prerecorded; store-and-forward) of information transmission, and active computational modeling based on preemptive information storage. For example, prefetching methods may be used to restore archived images to workstations in anticipation of needs for historical case comparisons, patient management, or clinical problem solving [24]. On the other hand, shifting ground may require isochronous (real-time) telecommunications transmission, and conversation through passive computing channels [23].…”