New software developments for quality mesh generation and optimization from biomedical imaging data

Histopathology-based near-realistic arterial wall reconstruction of a patient-specific cerebral aneurysm for fluid-structure interaction studies

BACKGROUND AND OBJECTIVE

Cerebral aneurysms occur as balloon-like outpouchings in an artery, which commonly develop at the weak curved regions and bifurcations. When aneurysms are detected, understanding the risk of rupture is of immense clinical value for better patient management. Towards this, Fluid-Structure Interaction (FSI) studies can improve our understanding of the mechanics behind aneurysm initiation, progression, and rupture. Performing retrospective hemodynamic analysis using an accurate computational model that is closer to the actual biological milieu could yield clinically useful rupture risk predictions. Currently, the geometric model for the FSI studies rely on imaging the flow-domain using Computed Tomographic Angiography (CTA) or Digital Subtraction angiography (DSA), which limits accurate discerning of the vessel wall thickness. Histopathological information has always been ignored in accurately reconstructing the geometric model for the aneurysm. The present study combines both the shape information of the 3D lumen model (as it existed in vivo), which is accurately rendered through the CTA, in conjunction with the wall thickness information extracted from histo-pathological 2D images of the aneurysm. Furthermore, fluid-structure interaction (FSI) simulations are performed to understand the influence of patient-specific wall contribution towards rupture.

METHODS

A 3D geometric model of the blood-flow domain of an anterior communicating artery (ACoA) aneurysm is extracted from the CTA of a patient that was surgically clipped. After safely clipping the aneurysm, the fundus of the aneurysm beyond the clip was cut and extracted. This was carefully preserved and sliced to obtain the wall thickness variation of the hoop at various axial sections. This study proposes a novel methodology of combining multi-modal image data to geometrically render the 3D model of the Cerebral aneurysm. The wall thickness extracted from the histological 2D cross-sectional images of the aneurysm is encapsulated around the 3D lumen model obtained from CT Angiographic data. To this end, a wall thickness transfer algorithm is developed to accurately reconstruct the patient-specific aneurysm wall thickness variation for the FSI simulations.

RESULTS

The wall thickness transfer algorithm accurately combines both the blood flow domain from the CT angiography and the histopathological images involving the wall thickness heterogeneity for the aneurysm. The patient-specific wall thickness variation, as it existed in vivo, has a mean wall thickness of 0.553 mm with a standard deviation of 0.256 mm. Detailed FSI simulations were performed to study the role of the patient-specific wall thickness (PWT) model vis-a-vis the uniform wall thickness (UWT) model. It was observed that the maximum wall stress for the UWT model was 13.6 kPa, while it was substantially higher for the PWT model (48.4 kPa). The maximum wall displacement for the UWT model was 58.5μm, while it was 162μm for the PWT model. Similarly, the mean wall stress for the UWT model was 2.13 kPa, while for the PWT model, it was 8.43 kPa. The mean wall displacement for the PWT model was substantially higher than the UWT model (52.58μm against 16.47μm).

CONCLUSION

The rendered patient-specific aneurysm wall model with its thickness variation, as it existed in vivo was obtained. Comparing fluid-structure interaction (FSI) simulation results, between the patient-specific wall-lumen combined model against the uniform wall thickness model have clearly shown that there were significant differences (p< 0.05) in the distribution of the hemodynamic parameters. The percentage difference in mean wall displacement and associated wall stress was 69% and 75%, respectively. Corresponding numbers for maximum wall displacement and maximum wall stress are 64% and 72%, respectively. Patient-specific fluid-structure interaction simulations show that, the present approach is highly valuable, as it improves our understanding towards rupture potential analysis for the cerebral aneurysms.

Image-based biomechanical models of the musculoskeletal system

Finite element modeling is a precious tool for the investigation of the biomechanics of the musculoskeletal system. A key element for the development of anatomically accurate, state-of-the art finite element models is medical imaging. Indeed, the workflow for the generation of a finite element model includes steps which require the availability of medical images of the subject of interest: segmentation, which is the assignment of each voxel of the images to a specific material such as bone and cartilage, allowing for a three-dimensional reconstruction of the anatomy; meshing, which is the creation of the computational mesh necessary for the approximation of the equations describing the physics of the problem; assignment of the material properties to the various parts of the model, which can be estimated for example from quantitative computed tomography for the bone tissue and with other techniques (elastography, T1rho, and T2 mapping from magnetic resonance imaging) for soft tissues. This paper presents a brief overview of the techniques used for image segmentation, meshing, and assessing the mechanical properties of biological tissues, with focus on finite element models of the musculoskeletal system. Both consolidated methods and recent advances such as those based on artificial intelligence are described.

Simulation of Neurotransmitter Flow in Three Dimensional Model of Presynaptic Bouton

In this paper a geometrical model for simulation of the nerve impulses inside the presynaptic bouton is designed. The neurotransmitter flow is described by using partial differential equation with nonlinear term. The bouton is modeled as a distorted geosphere and the mitochondrion inside it as a highly modified cuboid. The quality of the mesh elements is examined. The changes of the amount of neurotransmitter during exocytosis are simulated.

Direct volume manipulation for visualizing intraoperative liver resection process

This paper introduces a new design and application for direct volume manipulation for visualizing the intraoperative liver resection process. So far, interactive volume deformation and resection have been independently handled due to the difficulty of representing elastic behavior of volumetric objects. Our framework models global shape editing and discontinuous local deformation by merging proxy geometry encoding and displacement mapping. A local-frame-based elastic model is presented to allow stable editing of the liver shape including bending and twisting while preserving the volume. Several tests using clinical CT data have confirmed the developed software and interface can represent the intraoperative state of liver and produce local views of reference vascular structures, which provides a "road map of vessels" that are key features when approaching occluded tumors during surgery.