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

One of the distinct characteristics in radiologists' reading ofmultiparametric prostate MR scans, using reporting systems such as PI-RADSv2.1, is to score individual types of MR modalities, T2-weighted,diffusion-weighted, and dynamic contrast-enhanced, and then combine theseimage-modality-specific scores using standardised decision rules to predict thelikelihood of clinically significant cancer. This work aims to demonstrate thatit is feasible for low-dimensional parametric models to model such decisionrules in the proposed Combiner networks, without compromising the accuracy ofpredicting radiologic labels: First, it is shown that either a linear mixturemodel or a nonlinear stacking model is sufficient to model PI-RADS decisionrules for localising prostate cancer. Second, parameters of these (generalised)linear models are proposed as hyperparameters, to weigh multiple networks thatindependently represent individual image modalities in the Combiner networktraining, as opposed to end-to-end modality ensemble. A HyperCombiner networkis developed to train a single image segmentation network that can beconditioned on these hyperparameters during inference, for much improvedefficiency. Experimental results based on data from 850 patients, for theapplication of automating radiologist labelling multi-parametric MR, comparethe proposed combiner networks with other commonly-adopted end-to-end networks.Using the added advantages of obtaining and interpreting the modality combiningrules, in terms of the linear weights or odds-ratios on individual imagemodalities, three clinical applications are presented for prostate cancersegmentation, including modality availability assessment, importancequantification and rule discovery.

Combiner and HyperCombiner Networks: Rules to Combine Multimodality MR Images for Prostate Cancer Localisation