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 is a critical task that plays a vital role indiagnosis, treatment planning, and disease monitoring. Accurate segmentation ofanatomical structures and abnormalities from medical images can aid in theearly detection and treatment of various diseases. In this paper, we addressthe local feature deficiency of the Transformer model by carefully re-designingthe self-attention map to produce accurate dense prediction in medical images.To this end, we first apply the wavelet transformation to decompose the inputfeature map into low-frequency (LF) and high-frequency (HF) subbands. The LFsegment is associated with coarse-grained features while the HF componentspreserve fine-grained features such as texture and edge information. Next, wereformulate the self-attention operation using the efficient Transformer toperform both spatial and context attention on top of the frequencyrepresentation. Furthermore, to intensify the importance of the boundaryinformation, we impose an additional attention map by creating a Gaussianpyramid on top of the HF components. Moreover, we propose a multi-scale contextenhancement block within skip connections to adaptively model inter-scaledependencies to overcome the semantic gap among stages of the encoder anddecoder modules. Throughout comprehensive experiments, we demonstrate theeffectiveness of our strategy on multi-organ and skin lesion segmentationbenchmarks. The implementation code will be available upon acceptance.\href{https://github.com/mindflow-institue/WaveFormer}{GitHub}.

Unlocking Fine-Grained Details with Wavelet-based High-Frequency Enhancement in Transformers