scispace - formally typeset
Search or ask a question
Journal ArticleDOI

Patient Specific Vascular Benchtop Models for Development and Validation of Medical Devices for Minimally Invasive Procedures

27 Sep 2016-Vol. 1, Iss: 03, pp 1640008
TL;DR: This study proposes several patient specific vascular benchtop models for the development and validation of a robotic catheter for transcatheter aortic valve implantation and demonstrates that the described design process provides virtual models that are accurately linked to the physical models.
Abstract: Using realistic benchtop models in early stages of device development can reduce time and efforts necessary to move the device to further testing. In this study, we propose several patient specific vascular benchtop models for the development and validation of a robotic catheter for transcatheter aortic valve implantation. The design and manufacturing of these models, and their properties are presented. Additionally, it is demonstrated that the described design process provides virtual models that are accurately linked to the physical models.

Summary (3 min read)

1. Introduction

  • Severe aortic stenosis occurs in 3.4 % of the general elderly (>=75 years) western population (European countries and North America) (n = 9723 subjects) [1].
  • About 290 000 elderly patients with severe aortic stenosis are transcatheter aortic valve implantation (TAVI) candidates, under the current indications [1].
  • Later on in the development cycle , the device is tested in live animals (animal study) which partially can predict the device behavior in humans depending on the validity of the animal model [9].
  • Therefore, man-made replicas of patient-specific human anatomy used as benchtop models in the preclinical testing phase, will increase safety and efficacy before proceeding to clinical trials [10].
  • After collecting the appropriate geometric data, the model components may be molded, extruded, or machined.

2. Materials and Methods

  • The robotic catheters developed and tested in CASCADE are described by Vander Poorten et al. in the general CASCADE paper of this special issue, and by Devreker et al. (2014) [18].
  • Combining all these requirements in one model was not possible.
  • For more information on the algorithms and their validation, the authors refer to the general CASCADE paper by Vander Poorten et al. in this special issue.
  • In what follows, the model that is designed and printed for use during static testing is called the ‘rigid’ model.

2.1. Image based design of benchtop models

  • Developing and testing the robotic catheterization for TAVI involved the following anatomical regions: the aortic root, the aortic arch and the ascending and descending aorta with the main branches, in which most of the interventional procedures take place.
  • Using a graph cut based segmentation tool, the entire vasculature was segmented.
  • On average, the difference between the automatic result and the ground truth was comparable to the differences found between several operators performing a segmentation (average overlap measure of 0.87 for the graph cut segmentation tool versus 0.88 for the manual segmentation).
  • These can be easily removed via manual editing tools [19].
  • In the Mimics Innovation Suite®, the stl model was hollowed to create a wall thickness for later 3D printing.

2.2. Rigid benchtop model

  • The region of the rigid model included only the aortic arch with the coronary arteries, without the aortic valve leaflets.
  • The model was 3D printed using a transparent Tusk material (stereolitography technology).
  • In order to provide a smooth surface, a cosmetic finish was applied to the model which imposes a requirement of minimum 2mm wall thickness of the model.
  • Material properties of the material were not measured quantitatively as these were anyway not realistic.
  • The transparency of the rigid model was very high, making it well suited for catheter tracking based on a vision system .

2.3. Deformable HeartPrintTM Flex benchtop model

  • A TAVI procedure deals with the positioning of an aortic valve in a dynamic environment (the aortic annulus of the beating heart and the aorta surrounded by lungs).
  • This was assessed both qualitatively through inspection by experts (similar behavior but not identical), and quantitatively (see below).
  • The deformable model was based on the same images as the rigid model, including a wider range of anatomy: the aortic valve with leaflets, left and right coronary artery, ascending aorta, arch, brachiocephalic artery, left common carotid artery, left subclavian artery, descending and abdominal aorta, mesenteric artery, left and right renal, celiac trunk, left and right common iliac arteries.
  • Flex material can be used to print compliant models with properties that are within the range of human arterial tissue properties, namely a distensibility of 2.2 x 10-3 to 7.3 x 10-3 mmHg-1 described in literature [20, 21].
  • The calcifications were segmented in the Mimics Innovation Suite® based on Hounsfield units (HU), and transformed to a mesh file.

2.4. Deformable silicon benchtop model

  • Flex material has a lot of benefits as described above, two issues remain, namely its limited transparency, and its rupture sensitivity.
  • Moreover, it should have increased robustness to withstand the load during the tests.
  • Flex material’s transparency is suboptimal compared to silicon transparency.
  • This multi-step method combines usage of the Mimics Innovation suite® for segmentation of CT data, 3D printing of patient specific shells, and vacuum casting of the silicon model.
  • The resulting model possesses high robustness and transparency .

2.5. Testing of the Intravascular imaging compatibility of the deformable models

  • When developing the robotic catheterization devices under investigation in the CASCADE project, the preoperative data is taken into account for navigation.
  • Since the vasculature is deformable and dynamic in nature, preoperatively acquired information has limited value during the intervention.
  • In addition, information about the inside of the vessel enhances the user or controller’s awareness of the catheter/vessel system which could lead to better navigation.
  • Further information on this topic can be found in the general CASCADE paper by Vander Poorten et al. in this special issue.
  • The black line within the model was the IVUS probe, held by the hand on the right side of the picture.

2.6. Testing of the mechanical properties of the deformable models

  • The mechanical characteristics of both deformable models were assessed quantitatively, and compared with aortic tissue mechanical characteristics.
  • Planar biaxial tensile tests were performed on the HeartPrintTM Flex and silicon material on a BioTester (CellScale, Waterloo, Canada).
  • For more information on these material models and other continuum-mechanical aspects, please refer to [27].

3. Results

  • All three benchtop models were tested to check their suitability for use in the testing environment for the catheter development in the CASCADE project.
  • The compatibility with IVUS, and the mechanical properties are presented below.

3.1. IVUS compatibility

  • When testing the IVUS imaging in the models, the HeartPrintTM Flex material appeared very clearly in the ultrasound image.
  • Also, the silicon model had a satisfactory level of contrast on the IVUS images as it was possible to use image processing algorithms to extract the inner vessel wall contour .
  • (Images provided by Stammatia Gianarou, Su-Lin Lee, Liang Zhao, Imperial College, London).

3.2. Mechanical properties characterization

  • The material properties of the rigid model were not measured quantitatively as these were not realistic.
  • For the mechanical properties of the used materials of both deformable models, the results of the parameter fitting procedure for each of the samples are shown in table 1, along with material parameters of a typical healthy human aorta as obtained from unpublished data as collected in [28].
  • The relatively high variability (st. dev) of the HGO parameters of the aorta are in the same range as what is generally found when testing biological tissue [29, 30].

4. Discussion

  • The developed physical benchtop models provide a good representation of the clinical reality from geometrical standpoint.
  • The 3D printing process used to create the deformable model in HeartPrintTM.
  • There were algorithms developed to translate the pre-operative geometric data in a hex-mesh.
  • In conclusion, the proposed design and manufacturing process provides an interesting solution for different stages of medical device development as the testing environment from Materialise includes both realistic physical and virtual components with an accurate link between them.
  • Nevertheless, there are still differences in mechanical characteristics between the benchtop models and the real aortic tissue.

Did you find this useful? Give us your feedback

Figures (13)

Content maybe subject to copyright    Report

Patient Specific Vascular Benchtop Models for Development and Validation
of Medical Devices for Minimally Invasive Procedures
Maryna Kvasnytsia
a
, Nele Famaey
b
, Michal Böhm
c
, Eva Verhoelst
a
a
Materialise NV, Technologielaan 15, 3001 Leuven, Belgium
E-mail: Eefje.Verhoelst@materialise.be
b
KU Leuven, Biomechanics Section, Celestijnenlaan 300C, 3001 Heverlee, Belgium
c
Materialise Czech Republic, Predlicka 460/22, 400 02 Usti nad Labem, Czech Republic
Using realistic benchtop models in early stages of device development can reduce time and efforts necessary to move the device
to further testing. In this study, we propose several patient specific vascular benchtop models for the development and validation
of a robotic catheter for transcatheter aortic valve implantation. The design and manufacturing of these models, and their
properties are presented. Additionally, it is demonstrated that the described design process provides virtual models that are
accurately linked to the physical models.
Keywords: 3D printing, cardiovascular, HeartPrint
TM
Flex, silicon, material characterization, bench testing.
1. Introduction
Severe aortic stenosis occurs in 3.4 % of the general elderly (>=75 years) western population (European countries
and North America) (n = 9723 subjects) [1]. In the US alone, about 85000 aortic valve replacements are performed
annually [2]. Surgical aortic valve replacement (AVR) is the current standard of care, but it has been estimated
that between 30% and 60% of patients do not undergo AVR, owing to advanced age, left ventricular dysfunction,
or the presence of multiple coexisting conditions. Since the introduction of TAVI, these patients are offered an
alternative [3, 4, 5]. About 290 000 elderly patients with severe aortic stenosis are transcatheter aortic valve
implantation (TAVI) candidates, under the current indications [1].
Despite good results in a highly selected patient population, TAVI remains a delicate procedure. Interventionalists
are facing several difficulties limiting valve deployment quality, and causing problems that might affect the
patient’s health, such as difficult visualization of the valve, limited amount of control during distant steering of the
guidewire and the delivery of the valve [2, 6].
A European research project, “CASCADE”, Cognitive AutonomouS CAtheter operating in Dynamic
Environments, has been set up in order to advance the treatment of cardiovascular diseases by providing a new
dexterous and intelligent (self-aware, self-exploring) robotic instrument, initially focusing on endovascular aortic
valve replacement. The project aims to develop a unified control framework for continuum robots that operate in
complex and deformable environments in the particular case of TAVI procedures.
During the development of new medical devices, such as robotic catheters as envisioned in the CASCADE
project, various tests are required at consecutive stages of the product development lifecycle (Figure 1), including
the evaluation of overall device performance, the characterization of its physical properties, and the
characterization of the device interaction with living tissues. All evaluation and testing is subdivided in several
stages. After invention and prototyping of the device, pre-clinical tests are performed. Those can be laboratory
tests, benchtop model tests, animal studies and/or computer simulations [7]. After the pre-clinical testing phase,
the device goes into a clinical testing phase in which clinical trials are organized. Based on the results of these
clinical trials, a regulatory decision is made, and the product can be launched. Once on the market, the device is
still monitored for its performance [8].
Fig. 1. New medical device development stages according to the FDA [8].

In the preclinical testing phase, first, biocompatibility tests are performed to evaluate the tendency of the device
to cause damage to living tissues. Later on in the development cycle (Figure 1), the device is tested in live animals
(animal study) which partially can predict the device behavior in humans depending on the validity of the animal
model [9]. When entering the clinical testing phase the device is evaluated in human patients (clinical trial) to
determine the ability of the device to perform its intended use, and to evaluate safety and efficacy (device
performance). The FDA now encourages using benchtop models rather than animal models in the pre-clinical
phase while developing and commercializing medical devices in an attempt to reduce costs, ethical concerns, and
reliability of the testing results [7]. Animal models can differ significantly in anatomy from humans. For clinical
device design, correct anatomy, size and accessibility to the site during intervention are important aspects in
testing. Therefore, man-made replicas of patient-specific human anatomy used as benchtop models in the
preclinical testing phase, will increase safety and efficacy before proceeding to clinical trials [10].
Animal studies and benchtop model testing represent a special type of tests known as simulated use testing.
The motivation for this particular name is the fact that the animal is a simulation of the actual use (human)
environment. Other forms of models for simulation testing include cadavers (both human and animal) and
computational models [8]. The use of computer modeling has the potential to streamline the design, assessment
and evaluation of medical devices. In addition, those computational models can make clinical trials more efficient
by focusing on the most critical parameters in determining safety and effectiveness. Using both virtual models and
realistic benchtop models in the early phases of device development can significantly reduce time and minimize
efforts necessary to move the device to animal or cadaver testing stages. It is important to maintain a link between
the physical and computational models, since physical models can represent a source of reproducible “real world
data” for validation of the computational model [11].
The requirements for the ideal model to be used as substitute for a live animal, an animal cadaver, or a human
cadaver in the testing of devices are quite complex. They comprise realistic geometry with a high level of
anatomical detail, and similarity of physical properties (mechanical, electrical, electromagnetic, thermal, chemical,
etc…) to those of humans. The models should be designed to work at body temperature in the presence of fluids.
For some applications, the model needs to have a high level of transparency to allow visual or camera tracking of
the device location in the model. Additionally, some experimental setups require the models to be compatible with
X-Ray, MRI, ultrasound or different imaging methods.
The geometric data needed for fabrication is typically obtained in two ways. The traditional approach is to
obtain data from literature on morphology, or to obtain data from cadaver measurements. Since such measurements
are performed manually this method is time-consuming and permits a large degree of error. A preferable method
is to obtain the geometric data directly from one or several patients by means of scanning with consequent
automated virtual measurements.
After collecting the appropriate geometric data, the model components may be molded, extruded, or machined.
The materials used for benchtop models development include hydrogel, silicon rubber, natural rubber, acrylic
polymers, ceramics, cements, wood, styrofoam, metal, actual human tissues, actual animal tissues, and any
combination thereof. For complex geometries, traditional manufacturing techniques may become cumbersome and
expensive. In these cases rapid prototyping techniques offer an alternative. For some complex anatomies
researchers combine 3D printing and other methods such as dip-spinning, vacuum-casting, and injection molding
[12-17].
Most of the cardiovascular benchtop models present in the market are silicon models with fixed geometry.
Customization of the models often requires significant costs. A non-exhaustive list of benchtop models
manufacturers includes Elastrat, UnitedBiologics, BDC Laboratories, Fain Biomedical, Biomedical Modeling, The
Chamberlain group, and SynDaver Labs. The soft tissue material used for the SynDaver Labs models is claimed
to be almost identical to an actual human aorta. As a drawback, the material must be stored in a solution to preserve
its longevity. Customization is possible, although a complete patient specific model is not possible in the soft tissue
material. Moreover, those models are not transparent [15]. As far as our knowledge reaches, no information is
disclosed about the material characterization of the commercially used materials for benchtop models and the
comparison with real arterial tissue.
In this study, different patient specific benchtop models are designed and manufactured in different materials.
Their suitability for use in both a development and a testing environment for robotic catheterization devices under
investigation in the CASCADE project is evaluated. The mechanical properties of the deformable model’s
materials are assessed quantitatively, and compared to the ones of aortic tissue.
2. Materials and Methods
The robotic catheters developed and tested in CASCADE are described by Vander Poorten et al. in the general
CASCADE paper of this special issue, and by Devreker et al. (2014) [18]. To validate the catheter developments
and the associated algorithms for steering and control, benchtop models were created. The requirements for these
models were the following: (1) accurate vessel anatomy, (2) realistic mechanical properties resulting in similar

behavior of both the catheter and the vessel in real situations, (3) the possibility to simulate risk prone areas such
as calcifications, (4) compatibility with novel catheters and sensors such as intravascular ultrasound sensors, (5)
transparency to allow for visual tracking of the catheter, and (6) robustness for several testing rounds. Combining
all these requirements in one model was not possible. Therefore, three types of patient specific models, for different
stages in the testing process, were designed and printed, namely a rigid one for use in a static testing environment,
and two different deformable ones for use in a dynamic testing environment.
The static environment allowed for testing different algorithms developed in CASCADE necessary for navigation
of the catheter, namely:
1. catheter shape estimation algorithm: a spline-based shape estimation interpolating shape between
multiple electromagnetic (EM) sensors
2. automatic registration algorithms between pre-operative data and intra-operative data based on both force
and position (electromagnetic sensor) data
3. automatic catheter steering algorithms based on a minimum-energy model
4. localization of the catheter based on flow sensing
5. catheter shape modeling based on machine learning
6. automatic catheter steering based on machine learning
7. catheter teleoperation and advanced guidance experiments
8. semi-automatic catheter steering via automatic centerline following by the tip with manual
insertion/rotation
For more information on the algorithms and their validation, we refer to the general CASCADE paper by Vander
Poorten et al. in this special issue. All this algorithm testing was done in a rigid benchtop model. Aside from its
correct and realistic anatomy, the other main requirement for this model was its transparency. This allowed for
monitoring the catheter movement by a vision system during validation of the algorithms.
Once the algorithms tested in the static environment were stabilized, the realistic behavior of the material was of
higher importance compared to the transparency. The geometric correct vessel model captured only partially the
necessary information about the interaction between the vessel and the catheter as it did not provide information
about the forces that are exerted. Therefore, in a next stage, testing was performed in a dynamic environment
simulating the real situation during TAVI. As a consequence, the benchtop model had to be deformable, with
mechanical properties as close as possible to the real situation. Hence, two types of deformable models were
designed and printed in an attempt to combine several requirements at once for testing of the above algorithms in
a dynamic environment, and for testing two extra algorithms, namely:
1. simultaneous catheter and environment modeling: 3D reconstruction based on intravascular ultrasound
(IVUS) sensing, and EM sensors
2. detection of side branches and anatomic landmarks based on IVUS
As for the above algorithms, more information can be found in the general CASCADE paper by Vander Poorten
et al. in this special issue.
In what follows, the model that is designed and printed for use during static testing is called the ‘rigid’ model. The
second and third model, which are used during the later stages of testing in a dynamic environment, are called the
HeartPrint
TM
Flex model (non-transparent), and the ‘silicon’ model (transparent). The models are scaled 1:1 with
the real size of the patient case, and have the following length dimensions: 24 cm (rigid model), 62 cm
(HeartPrint
TM
Flex model), and 40 cm (silicon model).
2.1. Image based design of benchtop models
Developing and testing the robotic catheterization for TAVI involved the following anatomical regions: the
aortic root, the aortic arch and the ascending and descending aorta with the main branches, in which most of the
interventional procedures take place. Based on a CT scan (resolution of 512x512x1179, with pixel size 0.5mm) of
one case (courtesy of Dr. Herbert De Praetere, UZ Leuven), the Mimics Innovation Suite® (Materialise, Leuven,
Belgium) was used to obtain the virtual 3D reconstructed aortic model, and to prepare it in a printable format.
Using a graph cut based segmentation tool, the entire vasculature was segmented. Semiautomatic tools such as
region grow and multislice mask edit were used to clean up the vessel mask (Figure 2). The graph cut based
segmentation process used for vasculature segmentation was validated on 36 patient cases (CT). On average, the
difference between the automatic result and the ground truth was comparable to the differences found between
several operators performing a segmentation (average overlap measure of 0.87 for the graph cut segmentation tool
versus 0.88 for the manual segmentation). The cases not showing a good overlap with the ground truth typically
had inclusion of spine bone structures in the resulting segmented model. These can be easily removed via manual
editing tools [19]. Furthermore, manual segmentation tools were applied to define the region of interest for the
model. The segmented vasculature was represented by a triangulated surface (STL file format). This surface mesh
can be transformed to a volume mesh for later finite element analysis (FEA). In the Mimics Innovation Suite®,
the stl model was hollowed to create a wall thickness for later 3D printing. Specially designed connectors were
added to all models in order to connect the model to a (pumping) circulation system [20].

Fig. 2. CT images of one case (courtesy of Dr. Herbert De Praetere, UZ Leuven) with the vascular segmentation mask in
Mimics 16.0 (Materialise NV, Belgium).
2.2. Rigid benchtop model
The region of the rigid model included only the aortic arch with the coronary arteries, without the aortic valve
leaflets. The model was 3D printed using a transparent Tusk material (stereolitography technology). In order to
provide a smooth surface, a cosmetic finish was applied to the model which imposes a requirement of minimum
2mm wall thickness of the model. In order to apply this cosmetic finish from inside the model, it was printed in
two parts and then glued. Material properties of the material were not measured quantitatively as these were
anyway not realistic. The transparency of the rigid model was very high, making it well suited for catheter tracking
based on a vision system (Figure 3).
Fig. 3. The highly transparent 3D printed rigid benchtop model with tubes from the testing environment attached.
2.3. Deformable HeartPrint
TM
Flex benchtop model
A TAVI procedure deals with the positioning of an aortic valve in a dynamic environment (the aortic annulus of
the beating heart and the aorta surrounded by lungs).Therefore, the developed catheter and associated algorithms
should be tested in such a dynamic environment. This was created by designing and printing deformable models
which were connected to a pumping circuit to simulate the dynamics of the aorta during the procedure. As a
consequence, the model needed to have mechanical properties close to those of real arterial tissue. This was
assessed both qualitatively through inspection by experts (similar behavior but not identical), and quantitatively
(see below). A media file is available showing the deformation of the benchtop model incorporated into the
pumping circuit.

The deformable model was based on the same images as the rigid model, including a wider range of anatomy:
the aortic valve with leaflets, left and right coronary artery, ascending aorta, arch, brachiocephalic artery, left
common carotid artery, left subclavian artery, descending and abdominal aorta, mesenteric artery, left and right
renal, celiac trunk, left and right common iliac arteries. The deformable model was printed in HeartPrint
TM
Flex
with an Objet500 Connex printer (Stratasys) (Figure 4). Biglino et al. (2013) found that the HeartPrint
TM
Flex
material can be used to print compliant models with properties that are within the range of human arterial tissue
properties, namely a distensibility of 2.2 x 10
-3
to 7.3 x 10
-3
mmHg
-1
described in literature [20, 21]. The
distensibility for the HeartPrint
TM
Flex sample of 1.5 mm thickness is measured to be 3.0 x 10
-3
mmHg
-1
. For this
reason, the wall thickness of the deformable model was set to 1.5 mm as this material thickness guarantees realistic
mechanical properties of the model [20].
The deformable model was split into 4 parts to create a modular test-bed structure. Modularity allowed more
flexibility in terms of redesign of different anatomy types, and additionally limited the costs in case a particular
part would be damaged as it would only require rebuilding a part of the model. The main drawback of the
HeartPrint
TM
Flex material is its rupture sensitivity. The parts that were selected are:
Part 1: the aortic valve plus coronaries;
Part 2: the ventricular space;
Part 3: the arch and the thoracic aorta, at least 10 cm of the brachiocephalic, left common carotid, left
common subclavian;
Part 4: the abdominal aorta, iliac bifurcation including the renal, mesenteric and celiac artery.
Fig. 4. The 3D printed HeartPrint
TM
Flex deformable benchtop model showing its modular structure consisting of 4
segments, and the set of connectors.
For the development of intelligent robotic catheters it is important to take into account the calcifications present
in the vessel, which may hinder the catheter navigation or lead to embolization due to calcification rupture. The
device being developed should be able to recognize calcifications to avoid risk during the procedure. Therefore, it
is important to model calcifications in benchtop models for next device validation iterations. The severity and
extent of mineralization can be derived from analyzing preprocessed images. The calcifications were segmented
in the Mimics Innovation Suite® based on Hounsfield units (HU), and transformed to a mesh file. This method is
similar to the one described by Ewe et al. (2011) [22]. The intensity values of the calcifications on the CT images
allowed for a global threshold value of 800 HU to segment them. The corresponding material properties were
assigned to the vessel walls and the calcifications (Figure 5).

Citations
More filters
Journal ArticleDOI
01 Nov 2019
TL;DR: A novel aortic phantom reconstructed from patient-specific data with variable wall compliance that can be tuned without recreating the phantom is presented, allowing for the first time a phantom with tunable compliance.
Abstract: Validation of computational models using in vitro phantoms is a nontrivial task, especially in the replication of the mechanical properties of the vessel walls, which varies with age and pathophysiological state. In this paper, we present a novel aortic phantom reconstructed from patient-specific data with variable wall compliance that can be tuned without recreating the phantom. The three-dimensional (3D) geometry of an aortic arch was retrieved from a computed tomography angiography scan. A rubber-like silicone phantom was manufactured and connected to a compliance chamber in order to tune its compliance. A lumped resistance was also coupled with the system. The compliance of the aortic arch model was validated using the Young's modulus and characterized further with respect to clinically relevant indicators. The silicone model demonstrates that compliance can be finely tuned with this system under pulsatile flow conditions. The phantom replicated values of compliance in the physiological range. Both, the pressure curves and the asymmetrical behavior of the expansion, are in agreement with the literature. This novel design approach allows obtaining for the first time a phantom with tunable compliance. Vascular phantoms designed and developed with the methodology proposed in this paper have high potential to be used in diverse conditions. Applications include training of physicians, pre-operative trials for complex interventions, testing of medical devices for cardiovascular diseases (CVDs), and comparative Magnetic-resonance-imaging (MRI)-based computational studies.

11 citations

Journal ArticleDOI
27 Sep 2016
TL;DR: Progress is reported on the development of the necessary technology to autonomously steer catheters through the vasculature in terms of catheter design, vessel reconstruction, catheter shape modeling, surgical skill analysis, decision making and control.
Abstract: Advances in miniaturized surgical instrumentation are key to less demanding and safer medical interventions. In cardiovascular procedures interventionalists turn towards catheter-based interventions, treating patients considered unfit for more invasive approaches. A positive outcome is not guaranteed. The risk for calcium dislodgement, tissue damage or even vessel rupture cannot be eliminated when instruments are maneuvered through fragile and diseased vessels. This paper reports on the progress made in terms of catheter design, vessel reconstruction, catheter shape modeling, surgical skill analysis, decision making and control. These efforts are geared towards the development of the necessary technology to autonomously steer catheters through the vasculature, a target of the EU-funded project Cognitive AutonomouS CAtheters operating in Dynamic Environments (CASCADE). Whereas autonomous placement of an aortic valve implant forms the ultimate and concrete goal, the technology of individual building blocks to reach such ambitious goal is expected to be much sooner impacting and assisting interventionalists in their daily clinical practice.

8 citations

Journal ArticleDOI
TL;DR: In this paper , the authors developed a realistic, auto-adaptive, and visually plausible simulator to predict vessels' global deformation induced by the robotic catheter's contact and cyclic heartbeat motion.

1 citations

References
More filters
Book
23 Mar 2000
TL;DR: In this paper, the authors introduce the concept of stress and balance principles for tensors and invariance of tensors in the context of Vectors and Tensors, and present a survey of the main aspects of objectivity.
Abstract: Introduction to Vectors and Tensors. Kinematics. The Concept of Stress. Balance Principles. Some Aspects of Objectivity. Hyperelastic Materials. Thermodynamics of Materials. Variational Principles. References. Index.

2,082 citations

Journal ArticleDOI
TL;DR: A 2-year follow-up of patients in the PARTNER trial supports TAVR as an alternative to surgery in high-risk patients, but paravalvular regurgitation was more frequent after T AVR and was associated with increased late mortality.
Abstract: The rates of death from any cause were similar in the TAVR and surgery groups (hazard ratio with TAVR, 0.90; 95% confidence interval [CI], 0.71 to 1.15; P = 0.41) and at 2 years (Kaplan–Meier analysis) were 33.9% in the TAVR group and 35.0% in the surgery group (P = 0.78). The frequency of all strokes during follow-up did not differ significantly between the two groups (hazard ratio, 1.22; 95% CI, 0.67 to 2.23; P = 0.52). At 30 days, strokes were more frequent with TAVR than with surgical replacement (4.6% vs. 2.4%, P = 0.12); subsequently, there were 8 additional strokes in the TAVR group and 12 in the surgery group. Improvement in valve areas was similar with TAVR and surgical replacement and was maintained for 2 years. Paravalvular regurgitation was more frequent after TAVR (P<0.001), and even mild paravalvular regurgitation was associated with increased late mortality (P<0.001). Conclusions A 2-year follow-up of patients in the PARTNER trial supports TAVR as an alternative to surgery in high-risk patients. The two treatments were similar with respect to mortality, reduction in symptoms, and improved valve hemodynamics, but paravalvular regurgitation was more frequent after TAVR and was associated with increased late mortality. (Funded by Edwards Lifesciences; ClinicalTrials.gov number, NCT00530894.)

2,012 citations

Journal ArticleDOI
TL;DR: A structural continuum framework that is able to represent the dispersion of the collagen fibre orientation is developed and allows the development of a new hyperelastic free-energy function that is particularly suited for representing the anisotropic elastic properties of adventitial and intimal layers of arterial walls.
Abstract: Constitutive relations are fundamental to the solution of problems in continuum mechanics, and are required in the study of, for example, mechanically dominated clinical interventions involving soft biological tissues. Structural continuum constitutive models of arterial layers integrate information about the tissue morphology and therefore allow investigation of the interrelation between structure and function in response to mechanical loading. Collagen fibres are key ingredients in the structure of arteries. In the media (the middle layer of the artery wall) they are arranged in two helically distributed families with a small pitch and very little dispersion in their orientation (i.e. they are aligned quite close to the circumferential direction). By contrast, in the adventitial and intimal layers, the orientation of the collagen fibres is dispersed, as shown by polarized light microscopy of stained arterial tissue. As a result, continuum models that do not account for the dispersion are not able to capture accurately the stress–strain response of these layers. The purpose of this paper, therefore, is to develop a structural continuum framework that is able to represent the dispersion of the collagen fibre orientation. This then allows the development of a new hyperelastic free-energy function that is particularly suited for representing the anisotropic elastic properties of adventitial and intimal layers of arterial walls, and is a generalization of the fibre-reinforced structural model introduced by Holzapfel & Gasser (Holzapfel & Gasser 2001 Comput. Meth. Appl. Mech. Eng. 190, 4379–4403) and Holzapfel et al. (Holzapfel et al. 2000 J. Elast. 61, 1–48). The model incorporates an additional scalar structure parameter that characterizes the dispersed collagen orientation. An efficient finite element implementation of the model is then presented and numerical examples show that the dispersion of the orientation of collagen fibres in the adventitia of human iliac arteries has a significant effect on their mechanical response.

1,905 citations

Journal ArticleDOI
TL;DR: This article summarizes the methods and indices used to estimate arterial stiffness, and provides values from a survey of the literature, followed by recommendations of an international group of workers in the field who attended the First Consensus Conference on Arterial Stiffness, held in Paris during 2000.

1,096 citations

Journal ArticleDOI
TL;DR: The study showed the need to model nonstenotic human coronary arteries with nonatherosclerotic intimal thickening as a composite structure composed of three solid mechanically relevant layers with different mechanical properties.
Abstract: At autopsy, 13 nonstenotic human left anterior descending coronary arteries [71.5 ± 7.3 (mean ± SD) yr old] were harvested, and related anamnesis was documented. Preconditioned prepared strips (n =...

900 citations

Frequently Asked Questions (2)
Q1. What are the contributions mentioned in the paper "Patient specific vascular benchtop models for development and validation of medical devices for minimally invasive procedures" ?

In this study, the authors propose several patient specific vascular benchtop models for the development and validation of a robotic catheter for transcatheter aortic valve implantation. Additionally, it is demonstrated that the described design process provides virtual models that are accurately linked to the physical models. 

Further work should focus on manufacturing deformable models which would combine advantages of both technologies on the way to improve realistic properties of benchtop models, considering also minimizing time and cost needed required for production.