Figure 7 - uploaded by Stéphane Cotin
Content may be subject to copyright.
From left to right: (a) skeleton generated semi-automatically from the patient CTA, (b) 3D vascular surface reconstructed from the skeleton and cross-section data, and (c) combined arterial + venous sides obtained from the same patient data set. Arteries are in red, veins in blue. One can see the complexity and level of detail of the reconstructed models, and understand the importance to have optimized representations for real-time processing during a simulation. 

From left to right: (a) skeleton generated semi-automatically from the patient CTA, (b) 3D vascular surface reconstructed from the skeleton and cross-section data, and (c) combined arterial + venous sides obtained from the same patient data set. Arteries are in red, veins in blue. One can see the complexity and level of detail of the reconstructed models, and understand the importance to have optimized representations for real-time processing during a simulation. 

Source publication
Article
Full-text available
Computer-based interactive medical simulation is a revolutionary technique for improving the effectiveness of medical practice while reducing risk exposure to patients. Although traditionally focused on training, such simulations could be used, in the near future, for planning and rehearsing complex interventions, or even for assisting physicians i...

Contexts in source publication

Context 1
... ' is a scalar such that 0 < ' ( 1. This force is added to the external forces f in the global coordinate system before solving the linear system, and it can be shown that it acts as a damping force, where ' relates to the damping coe#cient of the model. To simulate accurately a device such as a guidewire or catheter we use a number of beam elements ranging from 100 to 200 ( Figure 27). With 6 degrees of freedom per node, we need to solve a linear system with about 1,000 unknowns at every time step. Although it can be done using iterative methods, it quickly becomes a limitation when integrating constraint due to contacts with the anatomy. To improve the computational performance of the method we proposes the following ...
Context 2
... is a matrix of coe#cients obtained from the direction cosines of angles between the local and global coordinate systems. To model a wire-like structure, we serially link a series of beam elements (see Figure 27). As a result, for the entire structure the global sti"ness matrix K is computed by summing the contributions of each element, thus leading to the following equilibrium ...
Context 3
... serious retinal eye problems that require surgery are caused by problems with the vitreous, the transparent gel-like substance which fills the center cavity of the eye (see Figures 66 and 67). The vitreous is attached to the retina, the thin layer of tissue lining the back of the eye. The retina has two parts: the peripheral retina and the macula. The macula allows to see very fine detail while the peripheral retina gives us our side vision. The vitreous is most strongly attached to the retina at the sides of the eye. It is also attached to the optic nerve, the macula and the large retinal blood ...
Context 4
... model of the cerebrovascular system: the method described above has been applied to various data sets, in particular contrast-enhanced magnetic resonance images (MRA) and contrast- enhanced computed tomography images (CTA). With MRA data, it is possible to obtain a fully automatic segmentation and reconstruction of the main cerebral vessels. The skeleton obtained from such data contains enough details for learning the basics of diagnostic angiography through catheter navigation and contrast agent injection. Using a higher resolution CTA data set, with manual interactions, it was possible to accurately segment and reconstruct a very detailed model of the vascular system, with about 3,000 arteries and 3,000 veins. Figure 7 shows the skeleton and the reconstructed surface. The skeleton has 3,388 lines and the reconstructed surface 58,612 polygons. This level of details is often required to allow an accurate diagnosis of the smallest brain vessels. From the same CTA, the venous system has also been reconstructed using our method. The resulting skeleton has 1,660 lines and the reconstructed surface 28,322 polygons. The combination of the arterial side and the venous side, shown in Figure 7 provides an ideal anatomical model for learning all essential steps of a ...
Context 5
... model of the cerebrovascular system: the method described above has been applied to various data sets, in particular contrast-enhanced magnetic resonance images (MRA) and contrast- enhanced computed tomography images (CTA). With MRA data, it is possible to obtain a fully automatic segmentation and reconstruction of the main cerebral vessels. The skeleton obtained from such data contains enough details for learning the basics of diagnostic angiography through catheter navigation and contrast agent injection. Using a higher resolution CTA data set, with manual interactions, it was possible to accurately segment and reconstruct a very detailed model of the vascular system, with about 3,000 arteries and 3,000 veins. Figure 7 shows the skeleton and the reconstructed surface. The skeleton has 3,388 lines and the reconstructed surface 58,612 polygons. This level of details is often required to allow an accurate diagnosis of the smallest brain vessels. From the same CTA, the venous system has also been reconstructed using our method. The resulting skeleton has 1,660 lines and the reconstructed surface 28,322 polygons. The combination of the arterial side and the venous side, shown in Figure 7 provides an ideal anatomical model for learning all essential steps of a ...
Context 6
... new bifurcation tiling with end-segment-grouping and adjacent-quadrant-grouping was also proposed to reduce surface patching artifacts when the branching angle is close to 0 or 90 degrees and to reduce unwanted surface twist. Final surface-fitting is answered by a Catmull-Clark or Loop subdivision algorithm ( Biermann et al., 2000) on the coarse mesh to improve surface smoothness. At each subdivision the number of triangles in the surface mesh increases by a factor 4, but only two levels of subdivision are typically needed to achieve excellent results (see Figure 7) since at each subdivision the cross-section is approximated more and more closely, thus giving an overall more accurate surface description. The subdivision also naturally leads to a smoother surface, which is key for simulating the navigation of medical ...

Citations

... In this method, the instrument is considered as a network of masses connected to each other by springs/dampers (Fig. 3). 12,23,44,45,[71][72][73][74] The springs not only give flexibility to the model but also constrain the distance between masses. Thus, the number of springs influences the behavior of the model. ...
... Thus, the number of springs influences the behavior of the model. 74 The deformable properties of the instrument depend on the parameters of the masses, springs, and dampers as follows: ...
... The mechanical properties of a guidewire/catheter change along the length, more flexibility at the distal side and more stiffness at the proximal side. Due to this property, some studies came with the idea of applying hybrid models, which means using either a combination of different techniques to model different parts of the instrument 10,11,28,74, or a new approach that was inspired by different models. 9,[87][88][89][90]91,92,93 In this way, they endeavored to make the simulation computationally more efficient. ...
Article
Full-text available
Guidewires and catheters are used during minimally invasive interventional procedures to traverse in vascular system and access the desired position. Computer models are increasingly being used to predict the behavior of these instruments. This information can be used to choose the right instrument for each case and increase the success rate of the procedure. Moreover, a designer can test the performance of instruments before the manufacturing phase. A precise model of the instrument is also useful for a training simulator. Therefore, to identify the strengths and weaknesses of different approaches used to model guidewires and catheters, a literature review of the existing techniques has been performed. The literature search was carried out in Google Scholar and Web of Science and limited to English for the period 1960 to 2017. For a computer model to be used in practice, it should be sufficiently realistic and, for some applications, real time. Therefore, we compared different modeling techniques with regard to these requirements, and the purposes of these models are reviewed. Important factors that influence the interaction between the instruments and the vascular wall are discussed. Finally, different ways used to evaluate and validate the models are described. We classified the developed models based on their formulation into finite-element method (FEM), mass-spring model (MSM), and rigid multibody links. Despite its numerical stability, FEM requires a very high computational effort. On the other hand, MSM is faster but there is a risk of numerical instability. The rigid multibody links method has a simple structure and is easy to implement. However, as the length of the instrument is increased, the model becomes slower. For the level of realism of the simulation, friction and collision were incorporated as the most influential forces applied to the instrument during the propagation within a vascular system. To evaluate the accuracy, most of the studies compared the simulation results with the outcome of physical experiments on a variety of phantom models, and only a limited number of studies have done face validity. Although a subset of the validated models is considered to be sufficiently accurate for the specific task for which they were developed and, therefore, are already being used in practice, these models are still under an ongoing development for improvement. Realism and computation time are two important requirements in catheter and guidewire modeling; however, the reviewed studies made a trade-off depending on the purpose of their model. Moreover, due to the complexity of the interaction with the vascular system, some assumptions have been made regarding the properties of both instruments and vascular system. Some validation studies have been reported but without a consistent experimental methodology.
... In addition, promising therapy for treating vascular diseases is interventional radiology; hence a guidewirecatheter combination and fluoroscopic guidance carry out more localized therapy to reduce recovery time for the patient when compared to traditional surgical procedures [30]. VR system also mocks contrast medium injections and simulates contrast medium washout in the vessel injected operations. ...
Article
Full-text available
This study proposes virtual-reality (VR) simulator system for double interventional cardiac catheterization (ICC) using fractional-order vascular access tracker and haptic force producer. An endoscope or a catheter for diagnosis and surgery of cardiovascular disease has been commonly used in minimally invasive surgery. It needs specific skills and experiences for young surgeons or postgraduate year (PGY) students to operate a Berman catheter and a pigtail catheter in the inside of the human body and requires avoiding damaging vessels. To improve the training in inserting catheters, a double-catheter mechanism is designed for the ICC procedures. A fractional-order vascular access tracker is used to trace the senior surgeons' consoled trajectories and transmit the frictional feedback and visual feedback during the insertion of catheters. Based on the clinical feeling through the aortic arch, vein into the ventricle, or tortuous blood vessels, haptic force producer is used to mock the elasticity of the vessel wall using voice coil motors (VCMs). The VR establishment with surgeons' consoled vessel trajectories and hand feeling is achieved, and the experimental results show the effectiveness for the double ICC procedures.
Thesis
Full-text available
La miniaturisation des composants des objets manufacturés est de plus en plus importante. Afin de réduire les coûts de production, il est nécessaire de développer des solutions robotiques de micromanipulation et de micro-assemblage et d’améliorer, en particulier, la cadence de manipulation qui reste faible sur les systèmes actuels (<1 cycle par seconde). L’objet de cette thèse de doctorat est de proposer des méthodes permettant de lever ce verrou scientifique de la micromanipulation haute cadence. Nous proposons une approche originale basée sur (i) une architecture parallèle à plateforme configurable permettant l’intégration d’un préhenseur, directement dans la structure et (ii) la miniaturisation du robot manipulateur permettant de réduire le ratio de taille entre le robot et l’objet. La fabrication du robot miniature a nécessité d’adapter sa conception aux spécificités de l’échelle et notamment de substituer les liaisons mécaniques conventionnelles par des articulations souples en polymère (PDMS). Nous montrons que cette approche permet de réduire les masses en mouvement et d’atteindre des hautes cadences (>10 cycles par seconde).Le manuscrit présente plusieurs méthodes de modélisation pour des mécanismes miniatures à articulations souples avant d’introduire et d’étudier une structure robotique originale pour la micromanipulation à haute cadence. Ce micromanipulateur a été testé expérimentalement et démontre la capacité de manipuler des objets micrométriques en effectuant 12 cycles de prise-dépose par seconde. Enfin, cette structure robotique originale a été également mise en œuvre à l’échelle macroscopique et illustré par la réalisation d’un manipulateur pour le tri de déchets.