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

Early detection of melanoma is crucial for preventing severe complicationsand increasing the chances of successful treatment. Existing deep learningapproaches for melanoma skin lesion diagnosis are deemed black-box models, asthey omit the rationale behind the model prediction, compromising thetrustworthiness and acceptability of these diagnostic methods. Attempts toprovide concept-based explanations are based on post-hoc approaches, whichdepend on an additional model to derive interpretations. In this paper, wepropose an inherently interpretable framework to improve the interpretabilityof concept-based models by incorporating a hard attention mechanism and acoherence loss term to assure the visual coherence of concept activations bythe concept encoder, without requiring the supervision of additionalannotations. The proposed framework explains its decision in terms ofhuman-interpretable concepts and their respective contribution to the finalprediction, as well as a visual interpretation of the locations where theconcept is present in the image. Experiments on skin image datasets demonstratethat our method outperforms existing black-box and concept-based models forskin lesion classification.

Coherent Concept-based Explanations in Medical Image and Its Application to Skin Lesion Diagnosis