Experimental Research On Intelligent Human-machine Collaborative Robot-assisted Craniotomy System

Background and Objective Common diseases in neurosurgery include brain trauma,brain tumors,cerebrovascular diseases,etc.The most basic and effective treatment for these diseases is surgery,and the most common surgical operation is craniotomy.Traditional craniotomy is performed by manual work.The shape and size of the bone window are designed according to the position of the lesion,and then the skull is cut by electric drill and milling cutter.This process is time consuming and laborious,and requires concentration to maintain stability during holding drill,avoiding dural tears,sinus rupture,nerve or blood vessel damage,etc.With the development of robots and artificial intelligence(AI),they are increasingly used in the field of neurosurgery.Robot-assisted stereotactic neurosurgery has been widely used in clinical practice,but use of robots in craniotomy have rarely been reported by far.In this study,the intelligent human-machine collaborative craniotomy robot system called Cranibot was developed,and the concept of human-machine collaborative interaction control was proposed to improve the accuracy of skull drilling and milling,reduce the incidence of tissue damage such as dural tear,shorten the time of surgery and the physical exertion of the doctor.Methods Integrated these elements into a robotic system for craniotomy,which includes UR5 robots,control panels,dual six-dimensional force sensor terminals,medical craniotomy drill and power system,NDI’s Plorias infrared optical navigator,self-designed program of navigation and positioning,and visual graphical user interface(GUI).So that the robotic arm can realize coordinating interactive control with the doctor under the guidance of navigation.Combined with preoperative CT data of head and preoperative drilling points of plan,the movement of the arm was restricted in a precise safe field,so the drilling and milling of craniotomy can be performed accurately and safely under the guidance of a doctor.The robot system of craniotomy was experimentally verified with three different experimental objects(PVC skull model,isolated pig head and Bama miniature pig)to examine its true performance,efficiency,accuracy,safety and practicability.Since the skull was symmetrical,each group of surgery performed bilateral symmetrical craniotomy on the skull.The skull model or pig head of the experimental group was drilled and milled on one side by the robot system,and the control group was operated by the same neurosurgeon at the other symmetrical position.At the end of the experiment,a square bone flap was finally milled,then recorded the data of force feedback,position error,time spent on craniotomy,and incidence of tissue damage such as dura tear in the experimental group and the control group,respectively.The difference between the experimental group and the control group was compared by statistical analysis finally.Results The robot system developed for three different experimental objects(head models,isolated pig heads and in-situ mini-pigs)can successfully complete tasks of craniotomy.In addition,the robot arm,the navigation system,the control host,the force sensor,the drill and the milling cutter had stable connection,smooth operation,and good performance without abnormal start and stop.The force feedback result showed that the feeding forces of the three experimental objects were 60N,100N,50N,respectively,and the contact forces during milling of them were all about-2N.The average positional errors of the experimental groups of the three experimental objects were 1.87 mm,3.13 mm,and 3.26 mm,respectively,the time of craniotomy were 6.64 min,7.79 min,and 11.83 min,and the total tissue injury rate was 5.6%.While the average positional errors of the control groups of the three experimental subjects were 3.14mm,3.83mm,4.49mm,and the time of craniotomy were 8.06min,9.05min,26.10min,and the total tissue injury rate was 16.7%.Conclusion As the first craniotomy robotic system developed independently,Cranibot is fully qualified for craniotomy and can improve the accuracy and efficiency of it,and reduce the incidence of tissue damage such as dural tears.However,due to the limited sample size,statistical analysis conclude that the robotic system has no difference from artificial craniotomy in avoiding tissue damage.In terms of force feedback,the robot system can accurately give the doctor a sense of force,and adjust the feeding force of the robot arm and the human hand according to the feedback,giving a safe and stable force control during the surgery.