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

Recently, artificial intelligence technologies and algorithms have become amajor focus for advancements in treatment planning for radiation therapy. Asthese are starting to become incorporated into the clinical workflow, a majorconcern from clinicians is not whether the model is accurate, but whether themodel can express to a human operator when it does not know if its answer iscorrect. We propose to use Monte Carlo dropout (MCDO) and the bootstrapaggregation (bagging) technique on deep learning models to produce uncertaintyestimations for radiation therapy dose prediction. We show that both models arecapable of generating a reasonable uncertainty map, and, with our proposedscaling technique, creating interpretable uncertainties and bounds on theprediction and any relevant metrics. Performance-wise, bagging providesstatistically significant reduced loss value and errors in most of the metricsinvestigated in this study. The addition of bagging was able to further reduceerrors by another 0.34% for Dmean and 0.19% for Dmax, on average, when comparedto the baseline framework. Overall, the bagging framework providedsignificantly lower MAE of 2.62, as opposed to the baseline framework's MAE of2.87. The usefulness of bagging, from solely a performance standpoint, doeshighly depend on the problem and the acceptable predictive error, and its highupfront computational cost during training should be factored in to decidingwhether it is advantageous to use it. In terms of deployment with uncertaintyestimations turned on, both frameworks offer the same performance time of about12 seconds. As an ensemble-based metaheuristic, bagging can be used withexisting machine learning architectures to improve stability and performance,and MCDO can be applied to any deep learning models that have dropout as partof their architecture.

A comparison of Monte Carlo dropout and bootstrap aggregation on the  performance and uncertainty estimation in radiation therapy dose prediction  with deep learning neural networks