| As a new type of minimally invasive surgery,laparoendoscopic single-site surgery(LESS)has the advantages of single incision and less bleeding.LESS robot is a combination of LESS and robotic techniques.It further realizes hand-eye coordination,remote operation,image navigation and other functions.However,compared with the traditional MIS,surgeons need to perform the surgery indirectly through a LESS robot,which puts forward a higher requirement for the safe operation of the surgeon.This paper focuses on the safety operation through a LESS robot,mainly from the two aspects of "guiding in training" and "guiding after training",and proposes a guiding method of safety operation and an automatic evaluation method of surgical skills.The main work of this study is as follows:Firstly,in order to facilitate data acquisition and simulation,a training simulator based on a LESS robot slave-arm is developed.3D models of the instrument and environmental objects are modeled by SolidWorks and 3DMAX.The training simulator is built on Microsoft Visual platform based on C++.It has the functions of data acquisition,real-time simulation,force feedback and automatic evaluation using OpenHaptics library.Based on the training simulator,a series of training tasks such as trajectory tracking,peg transfer and needle passing are developed.Automated evaluation is based on kinematics analysis.Secondly,virtual guidance is added to the training tasks of LESS,and the geometric and physical models of virtual guidance are established,which can guide the surgeon to the standard path in the course of training task.The virtual fixture theory is added to the virtual guidance to transform the operation area into a complex space with various artificial force potential fields.The complex space can be divided into three parts: security domain,warning domain and danger domain.Through the comparison experiment,the performance result are compared between those with virtual guidance and those without virtual guidance,which shows that virtual guidance can help the trainer to reduce the average deviation and redundancy of the actual operation.Finally,a new automatic evaluation method for surgical skills is proposed,which can point out clearly the surgical action that the trainer performs relatively poorly in a training task after the completion of training.It includes three parts: template operation extraction,subtask recognition and similarity evaluation.Through the peg transfer task,the results of the automatic evaluation method are proved to be highly correlative with the manual scoring of the experts,and its objectivity,efficiency and pertinence are verified. |
