3D High-Resolution Cardiac Segmentation Reconstruction from 2D Views using Conditional Variational Autoencoders
This addresses the challenge of whole-heart coverage in cardiac imaging for patients who cannot undergo time-consuming 3D MR scans, though it is incremental as it builds on existing conditional variational autoencoder techniques.
The paper tackled the problem of reconstructing 3D high-resolution cardiac segmentations from limited 2D views, achieving an average Dice score of 87.92 ± 0.15 on 400 healthy volunteers and outperforming competing methods.
Accurate segmentation of heart structures imaged by cardiac MR is key for the quantitative analysis of pathology. High-resolution 3D MR sequences enable whole-heart structural imaging but are time-consuming, expensive to acquire and they often require long breath holds that are not suitable for patients. Consequently, multiplanar breath-hold 2D cine sequences are standard practice but are disadvantaged by lack of whole-heart coverage and low through-plane resolution. To address this, we propose a conditional variational autoencoder architecture able to learn a generative model of 3D high-resolution left ventricular (LV) segmentations which is conditioned on three 2D LV segmentations of one short-axis and two long-axis images. By only employing these three 2D segmentations, our model can efficiently reconstruct the 3D high-resolution LV segmentation of a subject. When evaluated on 400 unseen healthy volunteers, our model yielded an average Dice score of $87.92 \pm 0.15$ and outperformed competing architectures.