English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Hip and pelvic fractures are serious injuries with life-threateningcomplications. However, diagnostic errors of fractures in pelvic X-rays (PXRs)are very common, driving the demand for computer-aided diagnosis (CAD)solutions. A major challenge lies in the fact that fractures are localizedpatterns that require localized analyses. Unfortunately, the PXRs residing inhospital picture archiving and communication system do not typically specifyregion of interests. In this paper, we propose a two-stage hip and pelvicfracture detection method that executes localized fracture classification usingweakly supervised ROI mining. The first stage uses a large capacityfully-convolutional network, i.e., deep with high levels of abstraction, in amultiple instance learning setting to automatically mine probable true positiveand definite hard negative ROIs from the whole PXR in the training data. Thesecond stage trains a smaller capacity model, i.e., shallower and moregeneralizable, with the mined ROIs to perform localized analyses to classifyfractures. During inference, our method detects hip and pelvic fractures in onepass by chaining the probability outputs of the two stages together. Weevaluate our method on 4 410 PXRs, reporting an area under the ROC curve valueof 0.975, the highest among state-of-the-art fracture detection methods.Moreover, we show that our two-stage approach can perform comparably to humanphysicians (even outperforming emergency physicians and surgeons), in apreliminary reader study of 23 readers.

Weakly Supervised Universal Fracture Detection in Pelvic X-rays