Two radiology-related search engines surfaced over the holidays, and both improve on the poor image quality and relevance of regular search engines.
Two radiology-related search engines surfaced over the holidays, and both improve on the poor image quality and relevance of regular search engines.
Goldminer was announced by the American Roentgen Ray Society in November. And a 2006 RSNA exhibit called GoogleMIRC reappeared in December as Yottalook.
Both Goldminer and Yottalook search only peer-reviewed medical imaging sources on the Web, screening out thousands of nonrelevant references or images that might be retrieved from routine keyword searches.
"Goldminer is more like PubMed for images," said its developer Dr. Charles E. Kahn, chief of radiology informatics at the Medical College of Wisconsin.
Images published in radiology journals are clinical and educational treasure troves, but they are not always handy when needed.
Yottalook developer Dr. Khan M. Siddiqui said about a year ago that he and his colleagues noticed radiologists relying more and more on search engines to find information at the time of image interpretation.
"Web search was becoming a decision support tool," said Siddiqui, chief of imaging informatics and cardiac CT/MR imaging at the VA Maryland Health Care System.
(The name stems from the yotta, an International System of Units prefix denoting 10 to the 24th power, the largest prefix yet formally introduced. The earth weighs 5977 yottagrams, for instance.)
While both Yottalook and Goldminer allow filtering for imaging modality, patient age, and gender, each has unique features.
"Goldminer searches by meaning, not just strings of characters," Kahn said.
Google, for example, finds only one image of phakomatosis on the Web. Goldminer returns 75 images for phakomatosis, none of which actually has the word phakomatosis in its caption. The captions say neurofibromatosis or tuberous sclerosis,which are types of phakomatoses.
Goldminer has content from only five journals, plus the EuroRad.org Web site, whereas Yottalook looks at relevance of results to radiology and is not limited to specific journals. Therefore, an image from The New England Journal of Medicine will be returned in Yottalook.
Yottalook, which is based on Google's indexing technology, is also better at searching multiple terms, such as hip avn.
Handling of thumbnails, however, is the most obvious distinction between Goldminer and Yottalook. Goldminer includes thumbnail previews of each image, which can be clicked on to view full-sized images. It also displays the title of the article that the figure comes from, which can be clicked on to access full texts.
Yottalook currently cannot show thumbnails due to copyright issues.
"In the future, when we clarify journal permission, we can cache images at Yottalook and show thumbnails," Siddiqui said.
Yottalook is currently getting about 2000 hits per day. Goldminer, which just started tracking hits, had no figures available.
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.