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

TL;DR

This paper introduces a conditional flow variational autoencoder that improves controllable synthesis of virtual anatomical populations, specifically cardiac ventricles, by capturing variability and demographic characteristics more effectively.

Contribution

The paper presents a novel conditional flow VAE that enhances the flexibility of virtual population synthesis by integrating normalising flows into the cVAE framework.

Findings

Outperforms standard cVAE in synthesising cardiac ventricles

Achieves lower generalisation and specificity errors

Better preserves clinically relevant biomarkers

Abstract

The generation of virtual populations (VPs) of anatomy is essential for conducting in silico trials of medical devices. Typically, the generated VP should capture sufficient variability while remaining plausible and should reflect the specific characteristics and demographics of the patients observed in real populations. In several applications, it is desirable to synthesise virtual populations in a \textit{controlled} manner, where relevant covariates are used to conditionally synthesise virtual populations that fit a specific target population/characteristics. We propose to equip a conditional variational autoencoder (cVAE) with normalising flows to boost the flexibility and complexity of the approximate posterior learnt, leading to enhanced flexibility for controllable synthesis of VPs of anatomical structures. We demonstrate the performance of our conditional flow VAE using a data set of cardiac left ventricles acquired from 2360 patients, with associated demographic information and clinical measurements (used as covariates/conditional information). The results obtained indicate the superiority of the proposed method for conditional synthesis of virtual populations of cardiac left ventricles relative to a cVAE. Conditional synthesis performance was evaluated in terms of generalisation and specificity errors and in terms of the ability to preserve clinically relevant biomarkers in synthesised VPs, that is, the left ventricular blood pool and myocardial volume, relative to the real observed population.