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

Spatial attention has been introduced to convolutional neural networks (CNNs)for improving both their performance and interpretability in visual tasksincluding image classification. The essence of the spatial attention is tolearn a weight map which represents the relative importance of activationswithin the same layer or channel. All existing attention mechanisms are localattentions in the sense that weight maps are image-specific. However, in themedical field, there are cases that all the images should share the same weightmap because the set of images record the same kind of symptom related to thesame object and thereby share the same structural content. In this paper, wethus propose a novel global spatial attention mechanism in CNNs mainly formedical image classification. The global weight map is instantiated by adecision boundary between important pixels and unimportant pixels. And wepropose to realize the decision boundary by a binary classifier in which theintensities of all images at a pixel are the features of the pixel. The binaryclassification is integrated into an image classification CNN and is to beoptimized together with the CNN. Experiments on two medical image datasets andone facial expression dataset showed that with the proposed attention, not onlythe performance of four powerful CNNs which are GoogleNet, VGG, ResNet, andDenseNet can be improved, but also meaningful attended regions can be obtained,which is beneficial for understanding the content of images of a domain.

A Novel Global Spatial Attention Mechanism in Convolutional Neural Network for Medical Image Classification