- AI
- Molecular Imaging
- CT
- X-Ray
- Ultrasound
- MRI
- Facility Management
- Mammography
Data warehouse speeds case file retrieval
Physicians at Stanford University Medical Center actually enjoy hunting down appropriate teaching and research cases. The center has developed and implemented a data warehouse that has reduced the search time and increased the number of cases found.
Physicians at Stanford University Medical Center actually enjoy hunting down appropriate teaching and research cases. The center has developed and implemented a data warehouse that has reduced the search time and increased the number of cases found.
"There is immense value in clinical data repositories," said Dr. Daniel L. Rubin, informatics research scientist at the medical center.
The data warehouse, dubbed Radbank, was developed using open source tools. It brings together more than 1.8 million radiological reports and 270,000 pathology reports. Rubin described Radbank at the 2004 RSNA meeting.
To evaluate Radbank's effectiveness in retrieving teaching files, researchers compared the database with manual searches for such studies as porcelain gallbladder, nephrocalcinosis, and pathologically confirmed viral esophagitis cases.
Physicians accessing the Radbank database located 12 to 13 times as many teaching cases in a fraction of the time needed for manual searching, according to the investigators. In one instance, a radiologist manually searching for pathologically confirmed viral esophagitis found no cases compared with 55 cases discovered by Radbank.
Rubin and colleagues also evaluated the use Radbank in identifying cohorts for two research studies:
- patients who had pelvic ultrasound exams and subsequently had pathologic evidence for retained products of conception (RPOC)
- patients diagnosed with intraductal papillary mucinous tumor (IPMT) of the pancreas on CT
Regarding these research-related tasks, Radbank found four to five times as many cases as manual searching. Because researchers using Radbank had access to pathologic cases, they discovered that 75 of the RPOC cases had pathologic confirmation. The radiologists relegated to manual searching abandoned the attempt to correlate cases to pathology after only three cases.
Radiologists engaged in manual searches needed days to weeks to complete their searches, according to the study. Radbank queries took only minutes to write and seconds to run.
Further work is being done on making Radbank more user-friendly, by moving toward a Web-based interface and away from an SQL-query format, Rubin said.
For more information from the online Diagnostic Imaging archives:
Integrated implementation revamps information storage
PACS archive evolves to meet enterprise needs
New standards improve interhospital dialogue
Study Reaffirms Low Risk for csPCa with Biopsy Omission After Negative Prostate MRI
December 19th 2024In a new study involving nearly 600 biopsy-naïve men, researchers found that only 4 percent of those with negative prostate MRI had clinically significant prostate cancer after three years of active monitoring.
Study Examines Impact of Deep Learning on Fast MRI Protocols for Knee Pain
December 17th 2024Ten-minute and five-minute knee MRI exams with compressed sequences facilitated by deep learning offered nearly equivalent sensitivity and specificity as an 18-minute conventional MRI knee exam, according to research presented recently at the RSNA conference.
Can Radiomics Bolster Low-Dose CT Prognostic Assessment for High-Risk Lung Adenocarcinoma?
December 16th 2024A CT-based radiomic model offered over 10 percent higher specificity and positive predictive value for high-risk lung adenocarcinoma in comparison to a radiographic model, according to external validation testing in a recent study.
