30 | Engineering Reality Magazine

S truct ure s

A-Z of stent simulation using Marc and how stents work inside

humans

Yashuhiro Shobayashi, Chief Technical Officer, IGA HealthcareKenji Mori, General Manager, R&D, Japan Life Line Co. Ltd.

Hiroshi Watanabe, Application Engineer, Hexagon | MSC SoftwareBhoomi Gadhia, Structures Product Marketing Manager, Hexagon | MSC Software

Volume XI - Summer 2020 | mscsoftware.com | 31

Introduction

The application of FEM to the medical field has been very successful in simulating hard tissues such as orthopedics and dental prosthesis areas. On the other hand, since soft tissues such as the heart are difficult to simulate, the mainstream of research in the medical field today is in vivo or in vitro. However, no matter how the molecular and cell biology evolves, it is not possible to carry out experiments that do not exist in nature. From that perspective, expectations for simulations are increasing.

Due to the development of finite element method technology and non-invasive measurement technology such as echo, CT and MRI in recent years, advanced research is progressing mainly in the academic field, and in silico animal experiments that replace conventional animal experiments is also becoming established. Considering that animal experiments will be restricted from the viewpoint of animal welfare in the future, it is necessary to establish the technology for in silico animal experiments as soon as possible.

Researchers often use in-house codes for analysis because they implement constitutive equations that reproduce the high anisotropy, self-contraction, and viscoelastic behavior of soft tissues. Marc is being implemented with an anisotropic hyperelastic model and a parallel rheological model. It is also highly anticipated that it can accurately perform complex contact analysis, which is a superiority of the past.

A typical medical device that can be analyzed by material modeling of soft tissue and complicated contact analysis is a stent. In this paper, we show the research to elucidate the movement of coronary arteries using 4D-CT and the development example of advanced stent.

Analysis for individual client to determine stent behavior under realistic working environment

In this project, we performed computational prediction from single component behavior to component behavior under real working environment. We used 4D CT-scan to capture individual client’s heart shape and its movement into digital data. 4D is basically 3-dimensional data plus the time history for that data. Once, the 4D scan data is captured, it is then converted into

FEA mesh and we apply boundary and load conditions to the FEA mesh. By doing this study, we were able to predict the organ behavior with high accuracy. With this organ behavior, we can predict stent behavior under realistic working environment.

The right coronary arteries (RCA) run on the surface of the epicardium and move in step with cardiac motion, expanding, contracting, bending, and twisting. These motions collectively make up the entire movement of the coronary artery. A detailed analysis of coronary motion is therefore essential to improve the results of percutaneous coronary intervention (PCI) for RCA ostial lesions. At present, 320-slice computed tomography (CT) provides 3-D coronary imaging during the cardiac cycle. In this study, we analyzed the motions of the RCA using finite element (FE) modeling of 4-D space-time data, a combination of 3-D space and 1-D time.

The CT imaging data obtained during a single cardiac cycle was divided into 20 datasets and stored in the Digital Imaging and Communications in Medicine (DICOM) format. The first DICOM dataset corresponds to the beginning of the systole. The heart image was segmented into regions (Figure 1A). Volume Graphisc (VG) technology is expected to improve the segmentation in the future. The CT volume data were converted into a polygonal image and, based on anatomic knowledge, each cardiac chamber was extracted from the image (Figure 1B). The resulting image consisted of core elements only: the aortic root, including the sinus of Valsalva, and the RCA (Figure 1C). The structure of the

RCA was modeled using beam elements, which have translational and rotational degrees of freedom. A shell element model was created based on the 15th DICOM dataset, corresponding to a time point between the early and atrial filling (Figure 1D). In this study Marc, a non-linear finite element analysis software suite, was utilized. The beam element model was placed near the central axis of the shell element model to measure the motion of the RCA. In the body the origin of the RCA moves with the cardiac motion, and the trajectory of this point was analyzed (Figure 2A). As shown in the enlarged view (Figure 2B), the RCA origin moves dynamically forward and backward in an elliptical pattern. In the next analysis, the aortic root was fixed as an immovable point at the RCA origin, and the motion of the RCA was calculated (Figure 2C; Movie S1).The proximal part of the RCA moves like a conical pendulum suspended from the RCA origin (Figure 2D; Movie S2). In addition, the coronary motion also affects the shape of the sinus of Valsalva (Movie S2). We estimated that this complicated movement would generate high mechanical stress on the stent placed in the ostial RCA. Stent fracture after DES implantation has recently become a major concern because of its potential association with in-stent restenosis and stent thrombosis. Use of this novel analytical method might have the potential to clarify the mechanism of stent fracture. For in vivo motion analysis of the coronary arteries, an imaging device must be able to visualize the motion of the coronary arteries throughout the cardiac cycle and also locate their positions in a coordinate system.

Figure 1 : (A) Segmented heart image obtained on 320-slice computed tomology. (B) Converted into a polygon image (C) Core element image of the aortic root and the right coronary artery (RCA). (D) Beam element model of RCA.

32 | Engineering Reality Magazine

The present study has shown that FE modeling of 4-D CT data can be used to successfully visualize and evaluate the detailed motion of the coronary arteries, although the potentially high radiation dose delivered in 4-D CT is a major concern. Further studies with a larger number of patients are needed to establish a more detailed motion analysis of the coronary arteries, including the LCA, to improve the outcomes of PCI.

To investigate the effect on the stent, we are investigating how tension, bending and twisting of the coronary arteries occur. Extract the movement of the heart with respect to the time axis by CT, about 20 frames per pulsation. Organize it as time series data. Attach the coronary arteries to the time-series data of the heart and simulate the movement. From the results, we detected which part of the coronary artery was pulled, which part was bent, and which part was twisted.

Biggest advantage using Marc for stent analysis

In following cases, Marc has a big advantage:

1. Analysis of the stent which is high in the rate of decline of the diameter.

2. With the contact with the elastic blood vessel.

3. Large-scale model (Complicated stent/vessel shape)

What kind of advantage is there in Marc?

• Automatic time step control (Auto Step) for large strain

- When large deformation occurs, the shape-memory alloy may show unstable behavior including the buckling. In this case, robustness automatic time step control feature is required. 

• Contact analysis feature for complicated contact analysis

- In the case of the contact analysis between the stent and elastic blood vessel, it is necessary to solve complicated contact.

• Parallel analysis feature for large-scale model

- A large-scale analysis model is necessary to express a detailed stent and vessel shape.

- The parallel analysis with DDM has a good scalability.

When we look at the stent Compression (in left picture), the stent is compressed in radial direction. Due to radial direction, high compression rate.

So, the way the stent behavior in blood vessel is analyzed is very realistically. First, the stent covered with tube is inserted into human

Figure 3 : Coronary artery behavior

body. Once the stent is properly placed in the blood vessel, the covering tube is removed from stent in the blood vessel. The stent springs back to it’s initial state and contact between stent and blood vessel. Marc predicts equilibrium state between stent and blood vessel.

Marc produced highly accurate results for the many stent analyses performed. For example, they use shape memory alloy behavior and we get confirmation that it works as intended. There was a feasibility study for mass production done where our customers noticed no crack, no failure due to highly deformed shape. We use Marc to accurately analyze many stent scenarios.

Figure 2 : (A) Trajectory of the right coronary artery (RCA) during a cardiac cycle. (B) Enlarged trajectory of the origin of the RCA. (C) Movement of the RCA when the aortic root is fixed. (D) The proximal end of the RCA moves like a conical pendulum.

Volume XI - Summer 2020 | mscsoftware.com | 33

Equivalent Von Mises stresses with collapsed stent in the artery

Stent placement in the artery

Equivalent Von Mises stresses with enlarged stent in the artery

Equivalent Von Mises stresses with the fully expanded stent in the artery