Xiaoshuai Chen

Design and evaluation of an encountered-type haptic interface using MR fluid for surgical simulators

A novel encountered-type haptic interface for surgical simulators is proposed. This interface has a container of MR (Magneto–Rheological) fluid, and an operator puts a surgical instrument into the fluid and can feel resistance force. The advantage of this interface is that an operator can move an instrument freely when it does not contact with MR fluid and change instruments easily. If an instrument is mounted mechanically on a haptic interface driven by servomotors, it is difficult to change surgical tools. On the other hand, the developed device does not require a procedure for changing tools and can increase a sense of reality. However, MR fluid cannot display large deformation of a tissue since its elastic region is small. Therefore, a container of the fluid is moved by servomotors. In this paper, concept and design of the interface and performance evaluations are described. In order to specify required display force, cutting force of a liver is analysed, and the maximum force is about 1.6 [N]. The device is designed based on this required force. Relationship between coil current and display force is measured, and the interface can exert 2.7 [N] when the current is 1 [A]. In addition, the validness of the proposed scheme using servomotors is evaluated.

Identification of mechanical properties of brain parenchyma for brain surgery haptic simulation

Dissection and removal of lesion area are fundamental operations in brain surgery. In order to correctly reproduce dissection and removal in brain surgery haptic simulation, mechanical properties of brain tissue must be investigated. In this work, porcine brains are used as specimens. Mechanical properties such as density, Poissons ratio, Youngs modulus, damping coefficient, and fracture stress of porcine brain parenchyma are measured and identified. Since haptic simulations require real-time computation of deformation and fracture of brain tissue, a linear finite element model and Rayleigh damping model are used. The Rayleigh damping coefficients are identified by solving optimization problems, so that the error between experimental results and simulation results is minimized.

Identification of physical properties of swine liver for surgical simulation using a dynamic deformation model

In recent medical field, surgical simulators with the technique of virtual reality are expected to provide a new means to support the surgical front. We have developed a simulation for brain surgery using a dynamic deformation model. Most of physics models, including the dynamic deformation model, are required to identify the physical properties of target tissues. In this research, we identified physical properties of swine liver. Youngs Modulus, Poissons Ratio and damping coefficient is necessary for the simulation. There are previous researches about identification of Youngs Modulus, but that about identification of Poissons Ratio and damping coefficient are few. Therefore, in this research, we conduct tension experiments to measure Youngs Modulus and Poissons Ratio, and vibration experiments to measure damping coefficient.

Experimental and numerical analysis of damage fracture mechanics of brain parenchyma

In this paper, a damage and fracture model of brain parenchyma is proposed for a haptic brain surgery simulation. It is assumed that microscopic damage begins by von Mises yield criterion, and the microscopic damage grows rate in proportion to volume strain. Tensile tests with two different strain rate were conducted using porcine brain parenchyma (tensile velocities: 0.1 (mm/s) and 1.0 (mm/s), mean length of specimens: 15 (mm)). Mechanical properties and proposed damage model parameters were identified by solving optimization problem with fitted curves of experimental data at 0.1 (mm/s). The proposed model and identified parameters were verified by comparing the simulation result and experimental data with tensile velocity of 1.0 (mm/s). A conventional damage model, simplified Lemaitre model, was implemented for comparison. Tensile simulations were performed with two models, proposed model and simplified Lemaitre model, with tensile velocity of 0.1 (mm/s). The simulation results were compared with the experimental result. It is confirmed that the proposed model well reproduce damage and fracture mechanics of brain parenchyma, while the simplified Lemaitre model could not well reproduce the ductility.

A viscoelastic model of brain parenchyma for haptic brain surgery simulations

This paper presents a viscoelastic model of brain parenchyma which is based on the generalized Maxwell model but considers also inertial force. The proposed model is implemented in finite element method (FEM). The viscoelastic behavior of brain parenchyma was confirmed by stress relaxation tests using porcine brain parenchyma. The stress relaxation characteristics were measured at strain 0.1, 0.2 and 0.3. The strains were applied by strain rates of 0.1 (s−1) and 1.0 (s−1). Viscoelastic parameters were identified by solving an optimization problem using experimental data and dynamic simulations. In order to verify the consistency and quality of the experimental data and the dynamic simulation, stress relaxation simulations were performed using the identified parameters.