Figure 1 - uploaded by Hessa Alfalahi
Content may be subject to copyright.
Pulmonary Vein Isolation (PVI) of the left atria. PVI is performed to cure drug-refractory Atrial Fibrillation (AF) patients. During the procedure, a catheter with an electrode on its tip is inserted to the left atria, whereby the Pulmonary Veins are circumferentially ablated, leading to the restoration of normal heart rhythm. A key success factor of the PVI is the achievement of transmural lesions that penetrate the whole thickness of the heart wall, thereby fully blocking the pulmonary veins. The rapid motion of the heart wall along the mitral valve axis represented in red precludes the accomplishment of constant contact force with the currently available ablation catheters.
Source publication
While the dilemma of motion tracking and force control in beating-heart surgery is previously addressed using active control architectures and rigid robotic actuators, this work leverages the highly controllable mechanical properties of concentric tube robots for intelligent, design-based force control in minimally invasive cardiac ablation. Briefl...
Contexts in source publication
Context 1
... ablation procedure is also called Pulmonary Vein Isolation (PVI), during which the pulmonary veins in the left atrium are circumferentially ablated. As shown in Figure 1, an ablation catheter with an electrode on its tip delivers radio-frequency energy to the entries of the pulmonary 4 The Biorobotics Institute, Scoula Superiore Sant'Anna, Pisa, Italy veins in a circumferential manner. This energy causes the transformation of the diseased tissue into scar tissue, thereby restoring the normal heart rhythm. ...
Context 2
... believe an ideal ablation catheter is that capable of maintaining a consistent, optimal contact force despite the motion of the heart tissue. This motion, based on the analysis in [15], is dominant along the mitral valve axis represented by the red dashed arrow in Figure 1. The average motion along this axis is 10.7 ± 2.7 mm. Accordingly, a 1 degree of freedom catheter tip is sufficient to follow the movement of the left atrial wall. ...
Context 3
... applied torque values were fitted to an interpolated function T(í µí¼), which has a single sinusoidal shape. Equation (18), plotted in Figure 10, was the best fit function with an accuracy of 0.21%, compared to the discrete torque values that resulted in the desired force, as per our simulation results. Our aim in the following section is to design and fabricate a proximal, compliant mechanism that will provide torque at the tubes' base, as close as possible to the interpolated torque function. ...
Context 4
... custom Hall Effect current sensor was fabricated to obtain real-time measurements of the current running through the tube. Regarding the fixture, we used the Computer Numerical Control (CNC) machine to create a slot in laminated wood, with the dimensions in table 1, as shown in Figure 11. ...
Context 5
... At í µí¼=0° and í µí¼=180°, the torque should be zero (see Figure 12(c) (left and right)) . Therefore, the moment arm of the tubes and at these two positions is equal to zero. ...
Context 6
... the moment arm of the tubes and at these two positions is equal to zero. • At í µí¼=90°, the torque should be zero (see Figure 12(c), middle). Therefore, we design the mechanism such that at 90°(straight tubes), the deformation of the spring is zero. ...
Context 7
... select the optimal design parameters of the proximal mechanism, we start by a theoretical representation of the system. The base assembly is shown in Figure 12. It consists of four gears; two upper driving gears that transmit torque to the two lower driven gears. ...
Context 8
... í µí»¼ is the angle of spring deflection and í µí°¾ is the torsional spring constant. Consequently, we formulate a constitutional relationship by defining the horizontal distance, í µí°¿, between the pins that attach the linkages to each of the driving gears (see Figure 12(b), middle). This is equivalent to the following: ...
Context 9
... í µí± is the horizontal distance from the center of the spring to the the center of the gears at the equilibrium spring state (middle), and í µí± is the vertical distance between the edge of the spring to the pin that attached the linkage to the gears (see Figure 12(b), middle). The distance, í µí°¿, is also directly related to the angle of tubes counter-rotation í µí¼, which is represented by the following function: ...
Context 10
... to the gears level, we define the force, í µí°¹ í µí± , that acts horizontally at the attachment point of the linkage (see Figure 12(b) (left)) as follows: ...
Context 11
... fine-tuned the dimensional parameters, especially í µí±, besides the torsional spring constant í µí°¾. The length, í µí±, however, is determined to be equivalent to the radius of the gears (see Figure 12(b) middle). The system of the equations was solved using standard nonlinear solver in MATLAB, with the input being í µí¼, and the outputs are í µí± í µí±í µí± , í µí»¼, í µí°¿, í µí°¹ í µí± and í µí± , as they are all related to í µí¼, either explicitly or implicitly as discussed earlier. ...
Context 12
... system of the equations was solved using standard nonlinear solver in MATLAB, with the input being í µí¼, and the outputs are í µí± í µí±í µí± , í µí»¼, í µí°¿, í µí°¹ í µí± and í µí± , as they are all related to í µí¼, either explicitly or implicitly as discussed earlier. Nonetheless, our objective is the torque at the tubes base, T, and Figure 13 shows the torque obtained by the simulated mechanism and that interpolated by equation (18). The spring constant, í µí°¾, is set to 2.4 Nm/rad, while í µí± is set to 0.08 m in the simulation. ...
Context 13
... vertebrae were formed using a Lulzbot TAZ 6 3D printer out of black Acrylonitrile Butadiene Styrene (ABS). The actual prototype is shown in Figure 14. We refer the reader to the physical prototype videos in the supplementary material to see its functionality principle https://drive.google.com/drive/folders/ ...
Context 14
... downward displacement of the force sensor, was incremented by 5 mm, at every step, and the force was measured and compared to its corresponding values obtained form the simulation of the torsion-less and the torsional models. Figure 15 shows the experimental setup. The aim of the conducted experiments is to prove the feasibility of the proposed prototype and to prove the success of our mathematical model, by comparing its simulation results to the experimental results. ...
Context 15
... aim of the conducted experiments is to prove the feasibility of the proposed prototype and to prove the success of our mathematical model, by comparing its simulation results to the experimental results. We have fabricated a phantom silicon tissue as in Figure 16. We have performed the experiments with two conditions, one trial with attaching the proximal spring to the base and one while the mechanism is completely removed from the base and the tubes are allowed to freely rotate without any proximal torque. ...
Context 16
... important remark is that this torsion is in fact what is causing the lower stiffness associated with case B, compared to case A, with both cases resulting a perfectly elastic tip. From Figure 17, it can be clearly noticed that the developed proximal mechanism is successful in flattening the force-displacement curve, and here it can be proved that by providing the right proximal torque obtained initially by our numerical analysis explained in section 3, the overall system's stiffness can be balanced resulting in a roughly zero stiffness tip and pseudo constant contact force. This also means that our proximal mechanism is capable of cancelling out the increasing stiffness of the bi-stable tubes system, maintaining constant potential energy over the tip's range of motion, which is referred to as static balance. ...
Context 17
... also means that our proximal mechanism is capable of cancelling out the increasing stiffness of the bi-stable tubes system, maintaining constant potential energy over the tip's range of motion, which is referred to as static balance. Figure 18 compares the mathematical model, without consideration of friction between the gears in the proximal mechanism, to the experimental results. To examine the impact of torsion, we also simulated our model with í µí± í µí± § = 0. To begin with, the force, which is the output in the simulation and the experiment always falls within the prescribed range as per the surgical prerequisites. ...
Context 18
... important determinant, with a huge impact on the catheter behavior is the vertical distance from the center of the torsional spring and the point of force application at the edge of the driving gears. To be more precise, this length is the moment arm of the torque around the center of the spring, as shown in Figure 19. Therefore, it represents an important design parameter for optimizing the proximal torque. ...
Context 19
... our experiments were performed considering both loading and unloading modes, and it has been observed that the developed proximal mechanism is associated with a remarkable amount of hysteresis, as illustrated in Figure 17, with the higher force level being under loading the tip with a downward force from the straight to the curved configuration. This is because the higher level of force occurs when the curvature is changing from straight to the downward curvature, injecting energy on the moving wall. ...
Context 20
... our eccentric tubes prototype, this energy loss takes place due to the gears in the proximal, compliant mechanism. Recall that rolling bearings are placed between the tubes and the vertebrae, eliminating frictional forces along the robotic backbone, as shown in figure 12(a). Upon torque transmission between the gears, the frictional forces between their teeth result in load-dependent energy dissipation [46]. ...
Context 21
... order to implement the load dependent efficiency in our model, we start by characterizing our gear system by three main discrete point of 90% efficiency. These three points are illustrated in Figure 12(b), which are í µí¼ = 0°, í µí¼ = 90°, and í µí¼ = 180° Therefore, the global efficiency of the system is the multiplication of these efficiencies at every pointing, yielding 72.9% in total. Recall that maximum energy dissipation occurs at the two mechanism configurations between í µí¼ = 0° and í µí¼ = 90°, and í µí¼ = 90° and í µí¼ = 180° (these configurations are not shown in Figure 12). ...
Context 22
... three points are illustrated in Figure 12(b), which are í µí¼ = 0°, í µí¼ = 90°, and í µí¼ = 180° Therefore, the global efficiency of the system is the multiplication of these efficiencies at every pointing, yielding 72.9% in total. Recall that maximum energy dissipation occurs at the two mechanism configurations between í µí¼ = 0° and í µí¼ = 90°, and í µí¼ = 90° and í µí¼ = 180° (these configurations are not shown in Figure 12). Accordingly, we improve the frictionless torque function taking into consideration the efficiency as follows: where í µí¼ í µí±í µí±í µí±¥ and í µí¼ í µí±í µí±í µí± are set to 0.729 and 0.271 respectively as they should be summing to 1. ...
Context 23
... we improve the frictionless torque function taking into consideration the efficiency as follows: where í µí¼ í µí±í µí±í µí±¥ and í µí¼ í µí±í µí±í µí± are set to 0.729 and 0.271 respectively as they should be summing to 1. When we implement this in our model, we recover the same experimental behavior as shown in Figure 21. ...
Context 24
... order to do so, we evaluate the partial derivatives of í µí°¹ with respect to L and D respectively as shown below. We then select one point from Figure 21, and we substitute with the values of í µí±, í µí± í µí± § , í µí¼ in the partial derivatives. We evaluate the derivatives for k=2.89 í µí± −1 , k í µí± § = −1.18m ...
Citations
... However, the drawbacks of electric heating are: (1) the requirement for a specific setup, which has some hardware limitations (2) tuning and calibrating the system to match the tube-specific transition temperature is challenging, and (3) the uneven thermal gradient/expansion along the tube can result in undesirable distorted shapes near the clamped ends [114]. Despite the high cost and inefficiency, the most commonly used approach is still the furnace heating technique, as recorded in 24 papers [35], [46], [62], [70], [81]- [83], [90], [120]- [135], as compared to the electric heating with only 6 papers [11], [114], [136]- [139], while both approaches are recorded in [80], [119], [140]. ...
Concentric tube robots (CTRs) have drawn significant research attention over the years, particularly due to their applications in minimally invasive surgery (MIS). Indeed, their small size, flexibility, and high dexterity enable several potential benefits for MIS. The research has led to an increasing number of discoveries and scientific breakthroughs in CTR design, fabrication, control, and applications. Numerous prototypes have emerged from different research groups, each with its own design and specifications. This survey paper provides an overview of the state of the art of the mechatronics aspects of CTRs, including approaches for the design and fabrication of the tubes, actuation unit, and end effector. In addition to the various hardware and associated fabrication methods, we propose to the research community, a unifying way of classifying CTRs based on their actuation unit architecture, as well as a set of specification details for the evaluation of future CTR prototypes. Finally, we also aim to highlight the current advancements, challenges, and perspectives of CTR design and fabrication.
Robotic surgery makes use of autonomous robots that can perform some surgical tasks on their own. Surgical robots performed well in conjunction with machine learning, particularly reinforcement learning (RL), allowing them to be used in complex environments, such as cutting a pre-determined pattern on soft tissue with surgical scissors and gripper. There is no doubt that soft tissue is deformable, so using a tensioning policy can determine appropriate tension direction from the pinch point at any time to have an accurate cut in the pre-determined trajectory. In this study, we used the deep reinforcement learning (DRL) approach to find an optimal tensioning policy for cutting soft tissues. In addition, we used an evolutionary algorithm with the operators appropriate to the problem and the learned tensioning policy to find the sequence of tensioning actions. The objective of this study is to determine the optimal tensioning policy and the best tensioning action sequence. The experimental results show that using the learned policy results in smaller damage and error and lead to the highest scores compared to the previous studies.
