1 July 2021 in Medical Image Analysis
We present a parametric physics-informed neural network for the simulation of personalised left-ventricular biomechanics. The neural network is constrained to the biophysical problem in two ways: (i) the network output is restricted to a subspace built from radial basis functions capturing characteristic deformations of left ventricles and (ii) the cost function used for training is the energy potential functional specifically tailored for hyperelastic, anisotropic, nearly-incompressible active materials. The radial bases are generated from the results of a nonlinear Finite Element model coupled with an anatomical shape model derived from high-resolution cardiac images. We show that, by coupling the neural network with a simplified circulation model, we can efficiently generate computationally inexpensive estimations of cardiac mechanics. Our model is 30 times faster than the reference Finite Element model used, including training time, while yielding satisfactory average errors in the predictions of ejection fraction (-3%), peak systolic pressure (7%), stroke work (4%) and myocardial strains (14%). This physics-informed neural network is well suited to efficiently augment cardiac images with functional data and to generate large sets of synthetic cases for training deep network classifiers while it provides efficient personalization to the specific patient of interest with a high level of detail.
19 June 2019 in Biomechanics and Modeling in Mechanobiology
12 April 2022 in Medical Image Analysis
We propose a differentiable volumetric mesh voxelization technique based on deformation of a shape-model, and demonstrate that it can be used to predict left-ventricular anatomies directly from magnetic resonance image slice data. The predicted anatomies are volumetric meshes suitable for direct inclusion in biophysical simulations. The proposed method can leverage existing (pixel-based) segmentation networks, and does not require any ground truth paired image and mesh training data. We demonstrate that this approach produces accurate predictions from few slices, and can combine information from images acquired in different views (e.g. fusing shape information from short axis and long axis slices). We demonstrate that the proposed method is several times faster than a state-of-the-art registration based method. Additionally, we show that our method can correct for slice misalignment, and is robust to incomplete and inaccurate input data. We further demonstrate that by fitting a mesh to every frame of 4D data we can determine ejection fraction, stroke volume and strain.
20 July 2022 in Computers & Fluids
28 December 2024 in Medical Image Analysis
Transthoracic Echocardiography (TTE) is a crucial tool for assessing cardiac morphology and function quickly and non-invasively without ionising radiation. However, the examination is subject to intra- and inter-user variability and recordings are often limited to 2D imaging and assessments of end-diastolic and end-systolic volumes. We have developed a novel, fully automated machine learning-based framework to generate a personalised 4D (3D plus time) model of the left ventricular (LV) blood pool with high temporal resolution. A 4D shape is reconstructed from specific 2D echocardiographic views employing deep neural networks, pretrained on a synthetic dataset, and fine-tuned in a self-supervised manner using a novel optimisation method for cross-sectional imaging data. No 3D ground truth is needed for model training. The generated digital twins enhance the interpretation of TTE data by providing a versatile tool for automated analysis of LV volume changes, localisation of infarct areas, and identification of new and clinically relevant biomarkers. Experiments are performed on a multicentre dataset that includes TTE exams of 144 patients with normal TTE and 314 patients with acute myocardial infarction (AMI). The novel biomarkers show a high predictive value for survival (area under the curve (AUC) of 0.82 for 1-year all-cause mortality), demonstrating that personalised 3D shape modelling has the potential to improve diagnostic accuracy and risk assessment.
15 May 2023 in NeuroImage: Clinical
To improve risk stratification in extracranial internal carotid artery disease (CAD), patients who would benefit maximally from revascularization must be identified. In cardiology, the fractional flow reserve (FFR) has become a reference standard for evaluating the functional severity of coronary artery stenosis, and noninvasive surrogates thereof relying on computational fluid dynamics (CFD) have been developed. Here, we present a CFD-based workflow using digital twins of patients' carotid bifurcations derived from computed tomography angiography for the noninvasive functional assessment of CAD. We reconstructed patient-specific digital twins of 37 carotid bifurcations. We implemented a CFD model using common carotid artery peak systolic velocity (PSV) acquired with Doppler ultrasound (DUS) as inlet boundary condition and a two-element Windkessel model as oulet boundary condition. The agreement between CFD and DUS on the PSV in the internal carotid artery (ICA) was then compared. The relative error for the agreement between DUS and CFD was 9% ± 20% and the intraclass correlation coefficient was 0.88. Furthermore, hyperemic simulations in a physiological range were feasible and unmasked markedly different pressure drops along two ICA stenoses with similar degree of narrowing under comparable ICA blood flow. Hereby, we lay the foundation for prospective studies on noninvasive CFD-based derivation of metrics similar to the FFR for the assessment of CAD.