Research On Soft Tissue Force Tactile Model And Reproduction Technique In Virtual Surgery

The virtual reality technology has brought continuous rapid development of related industries,including medicine,military,entertainment and so on.With the help of force feedback system,medical domain seeks breakthrough by combining itself with virtual reality technology,which brings new opportunities for trans geographical and cross material operation in telemedicine and surgery training.Virtual surgery has also emerged as the times require,enriching the content and essence of traditional surgery.In virtual surgery,modeling and simulation of soft tissue is the most important step.It not only determines the quality of virtual surgery,but also influences the realistic feelings and working efficiency of operators.Real time performance,simulation accuracy and telepresence are the most critical indicators of virtual surgery operation.Many experts and scholars have conducted in-depth research on the virtual soft tissue model.However,the real time and the simulation are always the paradox of the soft tissue deformation,which cannot be reconciled.In this paper,soft tissue simulation is studied on account of occasions and demands,and the corresponding soft tissue deformation simulation model is proposed.The main works of this paper are as follows:(1)A new virtual soft tissue torsion model is proposed based on the membrane analogy and Kriging interpolation.The membrane analogy method is used to compute the deformation of all the key sample points on the ridge.The Kriging interpolation method is utilized to compute the deformation of any points between the ridges.Firstly,using the Kriging interpolation method can accurately determine the deformed position of the points in the soft tissue deformation process.Secondly,the proposed membrane analogy method can stretch the membrane on a hole similar to the cross section of torque.It makes the computation and analysis of twist deformation become more intuitive,and improves the model accuracy under large-scale deformation.The integration of Kriging interpolation and membrane analogy method can accurately determine the twisting displacement after deformation.It can reduce the deformation time of the model and meet the real-time requirement of virtual interaction.(2)The coil spring model is proposed for simulating soft tissue deformation,which is composed of a plurality of coil springs connected in series.The superposition of torsion deformations produced by all coil springs is equivalent to soft tissue deformation on the surface.When the torsion force in a coil spring reaches its maximum,the next coil begins to generate torsion deformation.The force calculated during the torsion interaction under the torque force can be regarded as the force feedback in the real-time soft tissue deformation simulation.When force is applied to springs,same calculation method is utilized to calculate the torsion deformation of each coil spring,which relatively accelerates the calculation speed of spring deformation.By adjusting the axes radius,thickness,and cross-section width of coil spring,the system can simulate more kinds of soft tissue,and achieves wider applicability.(3)In view of the real-time and accuracy requirements of soft tissue grasping operation in virtual surgery,a spatial grasping deformation soft tissue model based on the BP neural network optimized by a genetic algorithm.The model divides soft tissue surface into grids,which generates displacement under the action of tension force.By studying the cylindrical spiral spring model,the relationship between force and displacement of grid points can be determined.The stress condition and deformation at any point on tissue surface can be obtained by training the data of grid points and tension force in the vertical direction with the optimized BP neural network.The model has good visual interaction and real-time feedback force signal,which can satisfy the demands of deformation simulation of soft tissue grasping operation in virtual surgery.