Wireless Capsule Endoscope Positioning Strategy Based On Magnetic Navigation

With the vigorous development of medical endoscopy,endoscopic diagnosis technology has made great progress,but cancer caused by various gastrointestinal lesions is still a challenge that researchers and medical workers need to face together.WCE has now become the focus of attention in the field of gastrointestinal disease detection and the development direction of future endoscopy technology.In order to improve the accuracy of lesion detection and the comfort of patient experience,WCE technology will gradually replace traditional gastrointestinal endoscopy in the future,and timely detection of lesions through effective screening will further reduce the probability of cancer.Therefore,the positioning research of WCE will provide strong support and help for the position detection of digestive tract lesions.In this paper,by exploring the positioning and attitude display of WCE in magnetic navigation technology,the position determination of the capsule when moving in the body was realized.When the doctor uses the mechanical arm to magnetically manipulate the WCE,it can better help the doctor determine the specific location of the lesion.In addition,for the difficulty in determining the motion direction of the WCE in the magnetic navigation animal experiment,the motion sensing module of the capsule was added,and the real-time attitude display of the capsule was successfully realized.In this paper,k-NN algorithm based on AI was used to locate WCE.Traditional positioning strategies for WCE were based on magnetic field but in this study,the positioning strategy based on electromagnetic wave which was RF to improve the positioning accuracy of the capsule.The research expounded the reasons for using k-NN algorithm,focused on deriving the mathematical model of k-NN algorithm,and introduced the classification basis and judgment of k-NN algorithm.A simulation environment was built based on the MATLAB tool,and the arbitrary motion path of the capsule was randomly generated to judge the error between the estimated capsule position and the actual capsule position.Further,in order to reduce the positioning error of WCE,the k-NN algorithm integrated with EKF was used to position capsules,and the mathematical model of EKF was added to MATLAB.The simulation results showed that the capsule positioning accuracy of the k-NN algorithm integrated with EKF was higher and the positioning error was smaller.The simulation will lay a sufficient and reliable theoretical foundation for future human experiment of magnetic navigation.In order to solve the problem of WCE attitude determination in the animal experiment of magnetic navigation,a motion-sensing module--MPU6050 was added to the structure of WCE designed in the future.In this paper,the module was used to detect the motion attitude of the capsule,and the attitude information of the capsule(including triaxial acceleration,gyroscope and euler angle)was collected in real time by the MATLAB host computer.Furthermore,the attitude modeling based on Quaternion was used to solve the attitude of the capsule and realize graphic visualization.In the tethered endoscope attitude test experiment,the attitude display of the endoscope was successfully realized,which laid a theoretical foundation and technical verification for the design of the WCE motion sensing module and the motion control of the magnetic navigation capsule in the future.