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

In this paper, we propose a novel semi-supervised learning (SSL) frameworknamed BoostMIS that combines adaptive pseudo labeling and informative activeannotation to unleash the potential of medical image SSL models: (1) BoostMIScan adaptively leverage the cluster assumption and consistency regularizationof the unlabeled data according to the current learning status. This strategycan adaptively generate one-hot "hard" labels converted from task modelpredictions for better task model training. (2) For the unselected unlabeledimages with low confidence, we introduce an Active learning (AL) algorithm tofind the informative samples as the annotation candidates by exploiting virtualadversarial perturbation and model's density-aware entropy. These informativecandidates are subsequently fed into the next training cycle for better SSLlabel propagation. Notably, the adaptive pseudo-labeling and informative activeannotation form a learning closed-loop that are mutually collaborative to boostmedical image SSL. To verify the effectiveness of the proposed method, wecollected a metastatic epidural spinal cord compression (MESCC) dataset thataims to optimize MESCC diagnosis and classification for improved specialistreferral and treatment. We conducted an extensive experimental study ofBoostMIS on MESCC and another public dataset COVIDx. The experimental resultsverify our framework's effectiveness and generalisability for different medicalimage datasets with a significant improvement over various state-of-the-artmethods.

BoostMIS: Boosting Medical Image Semi-supervised Learning with Adaptive Pseudo Labeling and Informative Active Annotation