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 Tools annual

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 monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Much progress has been made in the deep neural network (DNN) based diagnosisof mass lesions breast ultrasound (BUS) images. However, the non-mass lesion isless investigated because of the limited data. Based on the insight that massdata is sufficient and shares the same knowledge structure with non-mass dataof identifying the malignancy of a lesion based on the ultrasound image, wepropose a novel transfer learning framework to enhance the generalizability ofthe DNN model for non-mass BUS with the help of mass BUS. Specifically, wetrain a shared DNN with combined non-mass and mass data. With the prior ofdifferent marginal distributions in input and output space, we employ twodomain alignment strategies in the proposed transfer learning framework withthe insight of capturing domain-specific distribution to address the issue ofdomain shift. Moreover, we propose a cross-domain semantic-preserve datageneration module called CrossMix to recover the missing distribution betweennon-mass and mass data that is not presented in training data. Experimentalresults on an in-house dataset demonstrate that the DNN model trained withcombined data by our framework achieves a 10% improvement in AUC on themalignancy prediction task of non-mass BUS compared to training directly onnon-mass data.

Enhancing Non-mass Breast Ultrasound Cancer Classification With Knowledge Transfer