Elucidating the biomechanical behaviour of the myocardium is important for understanding cardiac physiology. This cannot be directly inferred from clinical imaging, but requires finite element (FE) simulations. However, conventional FE simulations are computationally expensive and often do not accurately reproduce clinically observed cardiac motions. Here, we propose the IMC-PINN-FE, a physics-informed neural network (PINN) framework that integrates imaged motion consistency (IMC) with FE modelling for patient-specific simulations of left ventricular (LV) biomechanics. Cardiac motion is first estimated from magnetic resonance imaging (MRI) or echocardiography using either a pre-trained attention-based network or an unsupervised cyclic-regularised network, followed by extraction of motion modes. IMC-PINN-FE then estimates myocardial stiffness and active tension by fitting clinical pressure measurements within seconds. Based on these parameters, IMC-PINN-FE performs FE modelling throughout the cardiac cycle and achieves an overall ∼ 75-fold speed-up over conventional FE simulations. Through motion constraints, its outputs of cardiac displacements across the cardiac cycle match imaged motions much more accurately, improving Dice coefficient from 0.849 to 0.927 compared to conventional FE methods. Concurrently, it preserves a realistic pressure–volume (P–V) relationship that is very similar to that of conventional FE. IMC-PINN-FE improves on existing PINN-FE models by enabling back-computation of myocardial properties and achieving better imaged motion fidelity. This strategy of using cardiac motions of a single individual to reconstruct shape modes avoids the need for a large dataset, and enables better patient-specificity. IMC-PINN-FE thus presents a robust approach for rapid, patient-specific, and image-consistent cardiac biomechanical modelling.

UNet and DT-VNet. Integrating TAM into SAM yields a temporal SAM that reduces Hausdorff distance (HD) from 3.99 mm to 3.51 mm on the CAMUS dataset, while integrating TAM into a pre-trained MedSAM reduces HD from 3.04 to 2.06 pixels after fine-tuning on the EchoNet-Dynamic dataset. On the ACDC 3D dataset, our TAM-UNet and TAM-DT-VNet achieve substantial reductions in HD, from 7.97 mm to 4.23 mm and 6.87 mm to 4.74 mm, respectively. Additionally, TAM's training does not require segmentation of ground truths from all time frames and can be achieved with sparse temporal annotation. TAM is thus a robust, generalizable, and adaptable solution for motion-awareness enhancement that is easily scaled from 2D to 3D. The code is available at https://github.com/kamruleee51/TAM.

Three-dimensional (3D) mesh reconstruction of the cardiac anatomy from medical images is useful for shape and motion measurements and biophysics simulations. However, 3D medical images are often acquired as 2D slices that are sparsely sampled (e.g., large slice spacing) and noisy, and 3D mesh reconstruction on such data is a challenging task. Traditional voxel-based approaches utilize non-differentiable pre- and post-processing that compromises fidelity to images, while mesh-level deep learning approaches require large 3D mesh annotations that are difficult to obtain. Differentiable cross-domain supervision from 2D images to 3D meshes is therefore crucial for enabling end-to-end optimization in medical imaging. While there have been attempts to approximate the voxelization and slicing of meshes that are being optimized, there has not yet been a method for directly using 2D slices to supervise 3D mesh reconstruction in a differentiable manner. Here, we propose a novel explicit differentiable voxelization and slicing (DVS) algorithm allowing gradient backpropagation to a 3D mesh from its slices, which facilitates refined mesh optimization directly supervised by the losses defined on 2D images. Further, we propose an innovative framework for extracting patient-specific left ventricle (LV) meshes from medical images by coupling DVS with a graph harmonic deformation (GHD) mesh morphing descriptor of cardiac shape that naturally preserves mesh quality and smoothness during optimization. The proposed framework achieves state-of-the-art performance in cardiac mesh reconstruction tasks from densely sampled (CT) as well as sparsely sampled (MRI stack with few slices) images, outperforming alternatives, including Marching Cubes, statistical shape models, algorithms with vertex-based mesh morphing algorithms and alternative methods for image-supervision of mesh reconstruction. Experimental results demonstrate that our method achieves an overall Dice score of 90% during a sparse fitting on multi-datasets. The proposed method can further quantify clinically useful parameters such as ejection fraction and global myocardial strains, closely matching the ground truth and outperforming the traditional voxel-based approach in sparse images.

Hemodynamics has a substantial influence on normal cardiovascular growth and disease formation, but requires time-consuming simulations to obtain. Deep Learning algorithms to rapidly predict hemodynamics parameters can be very useful, but their development is hindered by the lack of large dataset on anatomic geometries and associated fluid dynamics. This paper presents a new large-scale dataset of intracranial aneurysm (IA) geometries and hemodynamics to support the development of neural operators to solve geometry-dependent flow governing partial differential equations. The dataset includes 14,000 steady-flow cases and 200 pulsatile-flow cases simulated with computational fluid dynamics. All cases are computed using a laminar flow setup with more than 3 million cells. Boundary conditions are defined as a parabolic velocity profile with a realistic waveform over time at the inlet, and geometry-dependent mass flow split ratios at the two downstream outlets. The geometries are generated by a deep generative model trained on a cohort of 109 real IAs located at the middle cerebral artery bifurcation, capturing a wide range of geometric variations in both aneurysm sacs and parent vessels. Simulation results shows substantial influence of geometry on fluid forces and flow patterns. In addition to surface mesh files, the dataset provides volume data of velocity, pressure, and wall shear stresses (WSS). For transient cases, spatial and temporal gradients of velocity and pressure are also included. The dataset is tested with PointNet and graph U-Nets for WSS prediction, which showed relative L2 loss of 4.67\% for normalized WSS pattern.

Cardiac motion estimation is important for assessing the contractile health of the heart, and performing this in 3D can provide advantages due to the complex 3D geometry and motions of the heart. Deep learning image registration (DLIR) is a robust way to achieve cardiac motion estimation in echocardiography, providing speed and precision benefits, but DLIR in 3D echo remains challenging. Successful unsupervised 2D DLIR strategies are often not effective in 3D, and there have been few 3D echo DLIR implementations. Here, we propose a new spatial feedback attention (FBA) module to enhance unsupervised 3D DLIR and enable it. The module uses the results of initial registration to generate a co-attention map that describes remaining registration errors spatially and feeds this back to the DLIR to minimize such errors and improve self-supervision. We show that FBA improves a range of promising 3D DLIR designs, including networks with and without transformer enhancements, and that it can be applied to both fetal and adult 3D echo, suggesting that it can be widely and flexibly applied. We further find that the optimal 3D DLIR configuration is when FBA is combined with a spatial transformer and a DLIR backbone modified with spatial and channel attention, which outperforms existing 3D DLIR approaches. FBA's good performance suggests that spatial attention is a good way to enable scaling up from 2D DLIR to 3D and that a focus on the quality of the image after registration warping is a good way to enhance DLIR performance. Codes and data are available at: https://github.com/kamruleee51/Feedback_DLIR.

Embryonic heart development is a mechanosensitive process, where specific fluid forces are needed for the correct development, and abnormal mechanical stimuli can lead to malformations. It is thus important to understand the nature of embryonic heart fluid forces. However, the fluid dynamical behaviour close to the embryonic endocardial surface is very sensitive to the geometry and motion dynamics of fine-scale cardiac trabecular surface structures. Here, we conducted image-based computational fluid dynamics (CFD) simulations to quantify the fluid mechanics associated with the zebrafish embryonic heart trabeculae. To capture trabecular geometric and motion details, we used a fish line that expresses fluorescence at the endocardial cell membrane, and high resolution 3D confocal microscopy. Our endocardial wall shear stress (WSS) results were found to exceed those reported in existing literature, which were estimated using myocardial rather than endocardial boundaries. By conducting simulations of single intra-trabecular spaces under varied scenarios, where the translational or deformational motions (caused by contraction) were removed, we found that a squeeze flow effect was responsible for most of the WSS magnitude in the intra-trabecular spaces, rather than the shear interaction with the flow in the main ventricular chamber. We found that trabecular structures were responsible for the high spatial variability of the magnitude and oscillatory nature of WSS, and for reducing the endocardial deformational burden. We further found cells attached to the endocardium within the intra-trabecular spaces, which were likely embryonic hemogenic cells, whose presence increased endocardial WSS. Overall, our results suggested that a complex multi-component consideration of both anatomic features and motion dynamics were needed to quantify the trabeculated embryonic heart fluid mechanics.