Web-based image distribution is a popular choice among hospitals trying to extend PACS access beyond the radiology department. And the expense of equipping every department with sufficient PCs for hospital-wide viewing need not be prohibitive, according
Web-based image distribution is a popular choice among hospitals trying to extend PACS access beyond the radiology department. And the expense of equipping every department with sufficient PCs for hospital-wide viewing need not be prohibitive, according to researchers in Germany.
They found that even relatively unsophisticated hardware can meet performance targets, although few published studies discuss the performance requirements of PCs for Web-based image review. This information can be critical when purchasing IT hardware within strict budget constraints.
The researchers at the Johann Wolfgang Goethe University Hospital in Frankfurt, where Web-based viewing is already operational, decided to assess the influence of different PC setups on clinical workflow.
"The main focus of this research was to answer questions we were facing in our daily work," said Dr. Björn Bergh, who trained as a radiologist in Berlin before becoming IT director at the Frankfurt hospital.
Bergh and colleagues tested seven PCs covering three generations of processor (Pentium I, II, and III), using differing configurations of RAM, network, operating system, and graphic adaptor. They measured the time-to-display (TTD) CR, CT, and MR images from five separate examinations on each PC, while also altering screen resolution settings and switching from lossy to lossless compression.
A full analysis of the comprehensive study, which involved more than a half-million measurements, is published in the September issue of European Radiology.
Processor power proved to be the main limitation to performance. Doubling the processor speed halved the TTD, though even PCs with relatively modest processing power (350 MHz and above) managed to retrieve either one CR or 16 CT images within five seconds. Upping screen resolution beyond 1280 x 1024 pixels slowed the TTD most significantly for CR images.
Hospitals employing a reasonably regular program of IT hardware replacement should have no problems meeting the minimum threshold, Bergh said.
"The level of PC we are talking about here is not that high," he said. "We were actually astonished that you could use such a small PC."
But PC configuration is not the only factor involved in attaining adequate performance with Web-based viewing. A parallel study from the Frankfurt group on server requirements is forthcoming in European Radiology. Network speed can also influence clinical image TTD, Bergh said.
"Hospitals with a significantly slower network than the one we were using might find they have an information bottleneck in the network, not on the PC. So upgrading the PC will not help solve this problem," he said.
MRI-Based AI Radiomics Model Offers 'Robust' Prediction of Perineural Invasion in Prostate Cancer
July 26th 2024A model that combines MRI-based deep learning radiomics and clinical factors demonstrated an 84.8 percent ROC AUC and a 92.6 percent precision-recall AUC for predicting perineural invasion in prostate cancer cases.
Breast MRI Study Examines Common Factors with False Negatives and False Positives
July 24th 2024The absence of ipsilateral breast hypervascularity is three times more likely to be associated with false-negative findings on breast MRI and non-mass enhancement lesions have a 4.5-fold likelihood of being linked to false-positive results, according to new research.
Can Polyenergetic Reconstruction Help Resolve Streak Artifacts in Photon Counting CT?
July 22nd 2024New research looking at photon-counting computed tomography (PCCT) demonstrated significantly reduced variation and tracheal air density attenuation with polyenergetic reconstruction in contrast to monoenergetic reconstruction on chest CT.