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

Medical image segmentation can provide detailed information for clinicalanalysis which can be useful for scenarios where the detailed location of afinding is important. Knowing the location of disease can play a vital role intreatment and decision-making. Convolutional neural network (CNN) basedencoder-decoder techniques have advanced the performance of automated medicalimage segmentation systems. Several such CNN-based methodologies utilizetechniques such as spatial- and channel-wise attention to enhance performance.Another technique that has drawn attention in recent years is residual denseblocks (RDBs). The successive convolutional layers in densely connected blocksare capable of extracting diverse features with varied receptive fields andthus, enhancing performance. However, consecutive stacked convolutionaloperators may not necessarily generate features that facilitate theidentification of the target structures. In this paper, we propose aprogressive alternating attention network (PAANet). We develop progressivealternating attention dense (PAAD) blocks, which construct a guiding attentionmap (GAM) after every convolutional layer in the dense blocks using featuresfrom all scales. The GAM allows the following layers in the dense blocks tofocus on the spatial locations relevant to the target region. Every alternatePAAD block inverts the GAM to generate a reverse attention map which guidesensuing layers to extract boundary and edge-related information, refining thesegmentation process. Our experiments on three different biomedical imagesegmentation datasets exhibit that our PAANet achieves favourable performancewhen compared to other state-of-the-art methods.

PAANet: Progressive Alternating Attention for Automatic Medical Image Segmentation