For the first time, PACS image capture has started to migrate out of radiology into the enterprise."We're moving out of radiology and starting to use PACS to capture nonradiology images," said Dr. Gary Wendt, an associate professor of radiology at the
For the first time, PACS image capture has started to migrate out of radiology into the enterprise.
"We're moving out of radiology and starting to use PACS to capture nonradiology images," said Dr. Gary Wendt, an associate professor of radiology at the University of Wisconsin and director of Enterprise Image Integration.
PACS at the University of Wisconsin will soon house images generated by departments throughout the hospital - gastrointerology, pulmonary, and ENT - and visible light (VL) images. Cardiology uses the same PACS back end but has a separate archive due to different access needs, Wendt said.
Pilot projects have been run in GI and pulmonary, and trials have been completed in ENT. Equipment interfaces are in the final stages.
Motivation for the project comes from the desire to integrate images from different sources, resulting in better patient care.
"We began to wonder why we still had to have pathology slides found and brought over to clinical correlation conferences," Wendt said.
The idea is to get the pathology or endoscopy data online so at clinical correlation meetings clinicians can put the CT up next to the endoscopic image and correlate the two.
"There are a lot of images being generated in the enterprise - probably 60% to 70% of the image data - not being captured and made available. This compromises patient care," Wendt said.
The UW PACS enterprise image integration project will provide a common repository for images and make them accessible across the enterprise. First indications are that once images from pathology, ophthalmology, GI, pulmonary, and others are absorbed by PACS, actual radiology data will probably be only about 30% of the total image volume, Wendt said.
UW's PACS has been configured to handle the volume increase. First-tier storage consists of a 4-TB fiber channel RAID, although with 3:1 JPG lossless compression the effective capacity is 12 TB, or about a year.
Second-tier storage consists of 6-TB NAS (network attached storage) employing 10:1 lossy compression, yielding about 60 TB effective capacity or about five years. Deep archive is a pair of 300-TB AIT-3 jukeboxes, one onsite and one off.
A $200,000 servers upgrade resulted in multiple cluster servers (W2K) and a third set of servers offsite for disaster recovery. Web servers are clustered behind two load-balancing switches, one onsite, the other offsite.
The hospital's gigabit backbone and 100-Mb switched network is set to handle increases in traffic.
The effort is the first to incorporate VL image types into PACS.
"No one else is doing this that we know of," Wendt said. "VL is a brand-new spec and we're one of the first that has started using it."
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.