Abstract:Congenital heart disease (CHD) encompasses a spectrum of cardiovascular structural abnormalities, often requiring customized treatment plans for individual patients. Computational modeling and analysis of these unique cardiac anatomies can improve diagnosis and treatment planning and may ultimately lead to improved outcomes. Deep learning (DL) methods have demonstrated the potential to enable efficient treatment planning by automating cardiac segmentation and mesh construction for patients with normal cardiac anatomies. However, CHDs are often rare, making it challenging to acquire sufficiently large patient cohorts for training such DL models. Generative modeling of cardiac anatomies has the potential to fill this gap via the generation of virtual cohorts; however, prior approaches were largely designed for normal anatomies and cannot readily capture the significant topological variations seen in CHD patients. Therefore, we propose a type- and shape-disentangled generative approach suitable to capture the wide spectrum of cardiac anatomies observed in different CHD types and synthesize differently shaped cardiac anatomies that preserve the unique topology for specific CHD types. Our DL approach represents generic whole heart anatomies with CHD type-specific abnormalities implicitly using signed distance fields (SDF) based on CHD type diagnosis, which conveniently captures divergent anatomical variations across different types and represents meaningful intermediate CHD states. To capture the shape-specific variations, we then learn invertible deformations to morph the learned CHD type-specific anatomies and reconstruct patient-specific shapes. Our approach has the potential to augment the image-segmentation pairs for rarer CHD types for cardiac segmentation and generate cohorts of CHD cardiac meshes for computational simulation.

A Personalised 3D+t Mesh Generative Model for Unveiling Normal Heart Dynamics

Three-Dimensional Mesh Generation for Human Heart Model

Meshfree Implementation of Individualized Active Cardiac Dynamics.

Segmentation of Cardiac Tissues and Organs for CCTA Images Based on a Deep Learning Model

CHeart: A Conditional Spatio-Temporal Generative Model for Cardiac Anatomy

Learning Whole Heart Mesh Generation From Patient Images For Computational Simulations

Whole Heart Mesh Generation For Image-Based Computational Simulations By Learning Free-From Deformations

Spatio-temporal neural distance fields for conditional generative modeling of the heart

DeepMesh: Mesh-based Cardiac Motion Tracking using Deep Learning

A Deep Learning-Based Method for Rapid 3D Whole-Heart Modeling in Congenital Heart Disease

A Deep-Learning Approach For Direct Whole-Heart Mesh Reconstruction

Interpretable cardiac anatomy modeling using variational mesh autoencoders

Patient-specific Heart Geometry Modeling for Solid Biomechanics using Deep Learning

SDF4CHD: Generative Modeling of Cardiac Anatomies with Congenital Heart Defects

Generative Modelling of the Ageing Heart with Cross-Sectional Imaging and Clinical Data

LinFlo-Net: A Two-stage Deep Learning Method to Generate Simulation Ready Meshes of the Heart

A publicly available virtual cohort of four-chamber heart meshes for cardiac electro-mechanics simulations

Learning Interpretable Anatomical Features Through Deep Generative Models: Application to Cardiac Remodeling

Towards Integrated Noninvasive Personalization of Cardiac Electromechanics

Joint Segmentation and Motion Estimation of Left Ventricle with an Anisotropic Biomechanical Model

Meshless deformable models for 3D cardiac motion and strain analysis from tagged MRI