Modeling 3D cardiac contraction and relaxation with point cloud deformation networks
This work addresses the need for more detailed 3D cardiac analysis in medical imaging, offering a novel deep learning method that improves diagnostic accuracy for conditions like myocardial infarction, though it is incremental in applying point cloud techniques to this domain.
The authors tackled the problem of modeling 3D cardiac deformation, which is poorly captured by traditional clinical biomarkers, by proposing the Point Cloud Deformation Network (PCD-Net) to predict cardiac contraction and relaxation. They achieved average Chamfer distances below pixel resolution on over 10,000 cases and outperformed clinical benchmarks by up to 13% in MI detection and prediction tasks.
Global single-valued biomarkers of cardiac function typically used in clinical practice, such as ejection fraction, provide limited insight on the true 3D cardiac deformation process and hence, limit the understanding of both healthy and pathological cardiac mechanics. In this work, we propose the Point Cloud Deformation Network (PCD-Net) as a novel geometric deep learning approach to model 3D cardiac contraction and relaxation between the extreme ends of the cardiac cycle. It employs the recent advances in point cloud-based deep learning into an encoder-decoder structure, in order to enable efficient multi-scale feature learning directly on multi-class 3D point cloud representations of the cardiac anatomy. We evaluate our approach on a large dataset of over 10,000 cases from the UK Biobank study and find average Chamfer distances between the predicted and ground truth anatomies below the pixel resolution of the underlying image acquisition. Furthermore, we observe similar clinical metrics between predicted and ground truth populations and show that the PCD-Net can successfully capture subpopulation-specific differences between normal subjects and myocardial infarction (MI) patients. We then demonstrate that the learned 3D deformation patterns outperform multiple clinical benchmarks by 13% and 7% in terms of area under the receiver operating characteristic curve for the tasks of prevalent MI detection and incident MI prediction and by 7% in terms of Harrell's concordance index for MI survival analysis.