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

The generation of virtual populations (VPs) of anatomy is essential forconducting in silico trials of medical devices. Typically, the generated VPshould capture sufficient variability while remaining plausible and shouldreflect the specific characteristics and demographics of the patients observedin real populations. In several applications, it is desirable to synthesisevirtual populations in a \textit{controlled} manner, where relevant covariatesare used to conditionally synthesise virtual populations that fit a specifictarget population/characteristics. We propose to equip a conditionalvariational autoencoder (cVAE) with normalising flows to boost the flexibilityand complexity of the approximate posterior learnt, leading to enhancedflexibility for controllable synthesis of VPs of anatomical structures. Wedemonstrate the performance of our conditional flow VAE using a data set ofcardiac left ventricles acquired from 2360 patients, with associateddemographic information and clinical measurements (used ascovariates/conditional information). The results obtained indicate thesuperiority of the proposed method for conditional synthesis of virtualpopulations of cardiac left ventricles relative to a cVAE. Conditionalsynthesis performance was evaluated in terms of generalisation and specificityerrors and in terms of the ability to preserve clinically relevant biomarkersin synthesised VPs, that is, the left ventricular blood pool and myocardialvolume, relative to the real observed population.

A Conditional Flow Variational Autoencoder for Controllable Synthesis of Virtual Populations of Anatomy