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/

Zendy Plus

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

Zendy Tools

Zendy Open

Training deep learning models on medical datasets that perform well for allclasses is a challenging task. It is often the case that a suboptimalperformance is obtained on some classes due to the natural class imbalanceissue that comes with medical data. An effective way to tackle this problem isby using targeted active learning, where we iteratively add data points to thetraining data that belong to the rare classes. However, existing activelearning methods are ineffective in targeting rare classes in medical datasets.In this work, we propose Clinical (targeted aCtive Learning for ImbalaNcedmedICal imAge cLassification) a framework that uses submodular mutualinformation functions as acquisition functions to mine critical data pointsfrom rare classes. We apply our framework to a wide-array of medical imagingdatasets on a variety of real-world class imbalance scenarios - namely, binaryimbalance and long-tail imbalance. We show that Clinical outperforms thestate-of-the-art active learning methods by acquiring a diverse set of datapoints that belong to the rare classes.

CLINICAL: Targeted Active Learning for Imbalanced Medical Image Classification