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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The accurate classification of mass lesions in the adrenal glands (adrenalmasses), detected with computed tomography (CT), is important for diagnosis andpatient management. Adrenal masses can be benign or malignant and benign masseshave varying prevalence. Classification methods based on convolutional neuralnetworks (CNNs) are the state-of-the-art in maximizing inter-class differencesin large medical imaging training datasets. The application of CNNs, to adrenalmasses is challenging due to large intra-class variations, large inter-classsimilarities and imbalanced training data due to the size of the mass lesions.We developed a deep multi-scale resemblance network (DMRN) to overcome theselimitations and leveraged paired CNNs to evaluate the intra-class similarities.We used multi-scale feature embedding to improve the inter-class separabilityby iteratively combining complementary information produced at different scalesof the input to create structured feature descriptors. We augmented thetraining data with randomly sampled paired adrenal masses to reduce theinfluence of imbalanced training data. We used 229 CT scans of patients withadrenal masses for evaluation. In a five-fold cross-validation, our method hadthe best results (89.52% in accuracy) when compared to the state-of-the-artmethods (p<0.05). We conducted a generalizability analysis of our method on theImageCLEF 2016 competition dataset for medical subfigure classification, whichconsists of a training set of 6,776 images and a test set of 4,166 imagesacross 30 classes. Our method achieved better classification performance(85.90% in accuracy) when compared to the existing methods and was competitivewhen compared with methods that require additional training data (1.47% lowerin accuracy). Our DMRN sub-classified adrenal masses on CT and was superior tostate-of-the-art approaches.

Deep Multi-Scale Resemblance Network for the Sub-class Differentiation of Adrenal Masses on Computed Tomography Images