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

Widely used traditional supervised deep learning methods require a largenumber of training samples but often fail to generalize on unseen datasets.Therefore, a more general application of any trained model is quite limited formedical imaging for clinical practice. Using separately trained models for eachunique lesion category or a unique patient population will require sufficientlylarge curated datasets, which is not practical to use in a real-world clinicalset-up. Few-shot learning approaches can not only minimize the need for anenormous number of reliable ground truth labels that are labour-intensive andexpensive but can also be used to model on a dataset coming from a newpopulation. To this end, we propose to exploit an optimization-based implicitmodel agnostic meta-learning (iMAML) algorithm under few-shot settings formedical image segmentation. Our approach can leverage the learned weights fromdiverse but small training samples to perform analysis on unseen datasets withhigh accuracy. We show that, unlike classical few-shot learning approaches, ourmethod improves generalization capability. To our knowledge, this is the firstwork that exploits iMAML for medical image segmentation and explores thestrength of the model on scenarios such as meta-training on unique and mixedinstances of lesion datasets. Our quantitative results on publicly availableskin and polyp datasets show that the proposed method outperforms the naivesupervised baseline model and two recent few-shot segmentation approaches bylarge margins. In addition, our iMAML approach shows an improvement of 2%-4% indice score compared to its counterpart MAML for most experiments.

Meta-learning with implicit gradients in a few-shot setting for medical image segmentation