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 microscopy image cell segmentation, it is common to train a deep neuralnetwork on source data, containing different types of microscopy images, andthen fine-tune it using a support set comprising a few randomly selected andannotated training target images. In this paper, we argue that the randomselection of unlabelled training target images to be annotated and included inthe support set may not enable an effective fine-tuning process, so we proposea new approach to optimise this image selection process. Our approach involvesa new scoring function to find informative unlabelled target images. Inparticular, we propose to measure the consistency in the model predictions ontarget images against specific data augmentations. However, we observe that themodel trained with source datasets does not reliably evaluate consistency ontarget images. To alleviate this problem, we propose novel self-supervisedpretext tasks to compute the scores of unlabelled target images. Finally, thetop few images with the least consistency scores are added to the support setfor oracle (i.e., expert) annotation and later used to fine-tune the model tothe target images. In our evaluations that involve the segmentation of fivedifferent types of cell images, we demonstrate promising results on severaltarget test sets compared to the random selection approach as well as otherselection approaches, such as Shannon's entropy and Monte-Carlo dropout.

Knowing What to Label for Few Shot Microscopy Image Cell Segmentation