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

This paper addresses the problem of few-shot skin disease classification byintroducing a novel approach called the Sub-Cluster-Aware Network (SCAN) thatenhances accuracy in diagnosing rare skin diseases. The key insight motivatingthe design of SCAN is the observation that skin disease images within a classoften exhibit multiple sub-clusters, characterized by distinct variations inappearance. To improve the performance of few-shot learning, we focus onlearning a high-quality feature encoder that captures the unique sub-clusteredrepresentations within each disease class, enabling better characterization offeature distributions. Specifically, SCAN follows a dual-branch framework,where the first branch learns class-wise features to distinguish different skindiseases, and the second branch aims to learn features which can effectivelypartition each class into several groups so as to preserve the sub-clusteredstructure within each class. To achieve the objective of the second branch, wepresent a cluster loss to learn image similarities via unsupervised clustering.To ensure that the samples in each sub-cluster are from the same class, wefurther design a purity loss to refine the unsupervised clustering results. Weevaluate the proposed approach on two public datasets for few-shot skin diseaseclassification. The experimental results validate that our frameworkoutperforms the state-of-the-art methods by around 2% to 5% in terms ofsensitivity, specificity, accuracy, and F1-score on the SD-198 and Derm7ptdatasets.

Dynamic Sub-Cluster-Aware Network for Few-Shot Skin Disease Classification