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

Prediction models for clinical outcomes may be developed using a sourcedataset and additionally applied to new settings. Towards model externalvalidation and model updating in the new setting, one procedure is modelmodification learning that involves the dual goals of recalibrating overallpredictions as well as revising individual feature effects. Modificationlearning generally requires the collection of an adequate sample of trueoutcome labels from the new setting, which is frequently an expensive andtime-consuming process, as it involves abstraction by human clinical experts.To reduce the abstraction burden for such new data collection, we propose aclass of designs based on original model scores and their associated outcomepredictions. We provide mathematical justification that the general predictivescore sampling class results in valid samples for analysis. Then, we focusattention specifically on a stratified sampling procedure that we callpredictive case control (PCC) sampling, which allows the dual modificationlearning goals to be achieved at a smaller sample size compared to simplerandom sampling (SRS). PCC sampling intentionally over-represents subjects withinformative scores, where we suggest using the D-optimality and Binary Entropyinformation functions to summarize sample information. For design evaluationwithin the PCC class, we provide a computational framework to estimate andvisualize empirical response surfaces of the proposed information functions. Wedemonstrate the benefit of using PCC designs for modification learning,relative to SRS, through Monte Carlo simulation. Finally, using radiologyreport data from the Lumbar Imaging with Reporting of Epidemiology (LIRE)study, we illustrate the application of PCC for new outcome label abstractionand subsequent modification learning across imaging modalities.

Predictive case control designs for modification learning