RSNA: Rads search for (& find) search engines

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon

CHICAGO—Radiology search engines and a particularly open-access specialty are catalyzing radiologists' expertise and accelerating their responsiveness to complex cases, according to a panel of radiologists speaking Nov. 30 at the 96th annual scientific meeting of the Radiological Society of North America (RSNA).

Speaking as shareholders and/or founders of their respective radiology search engines, the presenters signaled an era of radiology not only dominated by quicker access to more information, but also by a confluence of IT and radiology professionals developing search engines and programs designed to provide high quality automated education and decision-support for common and differential diagnoses., hosted by the American Roentgen Ray Society (ARRS), provides radiologists free access to articles and images from more than 260 peer-reviewed journals, explained Charles E. Kahn, MD, MS, of the Medical College of Wisconsin in Milwaukee. ARRS GoldMiner conducts content-based searches across seven core journals, including the Journal of the American College of Radiology, Radiology and Radiographics, with the ability to vet results based on sex, age and imaging modality. Kahn reported an average retrieval time of .041 seconds and a search accuracy of 86 percent. was launched by several radiologists in response to largely unusable search results on the web for radiologists with typical search engines, which were marked by irrelevant results, closed access and poor indexing of peer-reviewed journals. Yottalook aims to provide radiologists with relevant, image-based queries, according to one of the site's founders, Woojin Kim, MD, from the Hospital of the University of Pennsylvania in Philadelphia.

Woojin claimed that the site serves 300,000 users in 174 countries, with musculoskeletal representing the most common searches. Yottalook can search exclusively within radiology journals as well as all PubMed material, and for a fee the search engine can be integrated with PACS for radiology searches within particular providers.

Statdx, a decision-support software program designed to assist radiologists with "rapid deductive reasoning with difficult cases," responds to queries for diagnostic confirmations as well as differential diagnoses, according to H. Ric Harnsberger, MD, professor of radiology at the University of Utah and CEO of Amirsys, the program's developer. Harnsberger claimed that Statdx has achieved unexpectedly widespread use among radiology trainees, with all university residency programs in the U.S. utilizing the program.

Harnsberger estimated that radiologists have questions and considerable uncertainties in approximately 5 to 10 percent of cases. Statdx enables radiologists to search text and images from peer-reviewed journals to assist with common and differential diagnoses, while also providing charts, data and graphs to summarize evidence and offers refresher guidelines for interventional procedures.

These and other programs have attempted to fill a niche in radiology. As a technologically advanced medical specialty, radiology has drifted almost entirely away from paper. Contradictorily, radiologists continue to report low satisfaction with online searches, the presentations underscored, sowing an open field for precise search engines to meet increasing workloads and aggregating evidentiary findings.