介入式微型医疗装置电磁定位方法的可行性研究，郭旭东，颜国正，为了跟踪胃肠道内介入式微型医疗装置的位置，考虑到人体介质的非导磁性，提出了基于电磁场理论的定位方法。主要研究了两种类型的中国科技论又在线http://www.paper.edu.cnThe configuration of thc alternating clcctromagnctic systcm is shown in figurc 4. Thcoutputted signal from the induction coil is programmable amplified and filtered. The useful signalis Lhen extracted and wirelessly transmitted outside. The signal received by rF receiver is inallysent to a computer for calculating the position and orientation3. Experiments and resultsIn order to testify the accuracy of the magnetic localization methods, the relevantexperiments were made in an unshielded room. A measuring apparatus was developed to measurethe position and orientation(P&o) of the target. As shown in figure 5, the reference coordinatesystem denoted by o-xyz was set up in the measuring apparatus. The origin was the projection ofrotating center around z axis onto the glass plane. The position of the target is described by thecoordinates in the rcfcrencc coordinatc systcm and the oricntation is dctcrmincd bascd on fulcrangle. The positional coordinates and attitude angles of the target could be easily acquired by themeasuring apparatus Mounting the larget on the supporting pole of the measuring apparatus, itsP&o can be arbitrarily adjusted. The excitation coils in both localization methods are cylinder inshape, 40 mm in diameter and 8 mm in lengthIn the experiments, the real P&o were directly measured by means of the measuringapparatus. Meanwhile, the P&o were calculated by the localization method. Comparing thecalculated result with the real value. the measurement error could be obtainedGuiding poleGuiding Pole Induction CoilSupporting poleSupporting PoleMagnetiExcitation coilsExcitation coilGlass planeGlass planeFig 5 Quasi-static electromagnetic localizationig. 6 Alternating electromagnetic localizationThe following formula is used to estimate the errors of the position and the orientation∧x1,=py1, Aa=a-a, AB=B-B1,Ay=r-rAs for the quasi-static electromagnetic localization method, the orientation errors range from0.0140 to 0.3102(about 0. 8to 178)in the experiment. 65% of the samples have the orientationerrors smaller than 0.1500. The maximum error of the x, y and z component is 0.063 m,0.069 mand 0.059 m, respectively. 67%0, 65%o and 72%o samples have error of x, y and z componentsmaller than 0.030m. The average error of x, y and z component is 0.026m, 0.03lm and 0.022mrespeclivelyWhen the localization distance is between 0. 140m and o170m. the localization errors arelargcr than 0. 100m. As the distance is farcr than 0. 170m, the magnctic ficld gcncrated by thcexcitation coils is even less than the earth magnetic field. It is because that the increasing distancesharply reduces the intensity of the magnetic field. Low signal-to-noise ratio(SNR)results in thelarge errors. What's more, there's hardly any possiblility to use a larger excitation current becauseof heating effectAs for the alternating electromagnetic localization method, the orientation errors range from中国科技记文在线http://www.paper.edu.cn0.0021 to 0. 2632(about 0. 1to 15.1 )in the cxpcrimcnt. 76% of the samples havc the oricntationerrors within 0.1500. The maximum error of the x, y and z component is 0.050 m, 0.058 m and0.052 Im, respectively. 85%,80%and 86% samples have error of X, y and z component smallerthan 0.030m. The average error of x, y and z component is 0.012m, 0.019m and 0.01lmrespcctivcly. Thesc results arc satisfactory, in comparison with the RF mcthod uscd in M2A Thcerror of 87.0%o samples for the latter is smaller than 0.060 m and the biggest error is 0.114m[2]In the quasi-static electromagnetic localization method the localization distance is not morethan 0.17m. With the increasing distance, the localization accuracy decreases sharply. Besides, thislocalization method may experience interference from 1/f noise existing in the circuits and otherdevices that produce magnetic fields. In comparison, the alternating electromagnetic methodprovides an up to three orders better SNR, is not sensitive to dC errors or 1/f noise. Thelocalization distance ranges from 0.03m Lo 0.4m. However, the alternating magnelic field isintended to be distorted due to the eddy currents induced in nearby metals when the field ischanging. Thcrcforc mctal objects in thc vicinity of excitation coils may rcducc thc localizationaccuracv4. ConclusionsTwo varieties of electromagnetic localization methods for localizing in-vivo micro-deviceshave been invesligated in the paper. In order to study the leasibility of localizing in-vivo medicalmicro-devices with the two methods, relevant experiments have been performed. Comparisonbctwccn thc quasi-static clcctromagnctic localization and the alternating clcctromagnctic thc staticelectromagnetic localization has been madeThe experiments show that both varieties of electromagnetic localization methods succeed indetermining the position and orientation of a target. In comparison with the quasi-staticelectromagnetic method, the alternating electromagnetic method acquires higher accuracy, higherstability and wider measurement rangeacknowledgementsThis work was supported by Specialized Research Fund for the Doctoral Program of HigherEducation (No. 20040248033)and the Nalional Hi-Tech Research and Development Program(863)of China(No. 2006AA04Z368)References[1] Iddan g, Mcron g, Glukhovsky A. Swain P. Wirclcss capsule endoscopy Naturc. 2000; 405: 417-420[2] Fischer D, Schreiber R, Levi D, et al. Capsule endoscopy: The localization system. Gastrointestinal EndoscopyClinics of north america. 2004: 14: 25-31[3] Wang WX, Yan GZ, Sun F. A non-invasive method for gastrointestinal parameter monitoring. World JGastroenterol.2005;11(4):521-524[4] Scott Thomas, 2006, SmartPill redefines "noninvasive" Buffalo Physician, 40(3), 13-145 E Stathopoulos, V Schlageter, B. Meyrat, Y. Deribaupierre&P Kucera. Magnetic pill tracking: a novelnon-invasivc tool for invcstigation of human digestive motility. Ncurogastrocntcrol Motil, 2005. 17: 148-154[6 Paperno E, Keisar P, Three-Dimensional magnetic Tracking of Biaxial Sensors. IEEE Transactions onMagnetics.2004,40:1530-1537[7 Wang XN, Max Q.H. Meng. Study of a position and orientation tracking method for wireless capsuleendoscope. International Journal of Information Acquisition. 2005, 2(2): 1 13-121el1MagneticIct catalog. Availablehttp://www.ssec.honeywell.com/magnetic.AccessedAugust2006[9] Guo Xudong, Yan Guozheng, Hc Wenhui Modeling of consccutivc tracking systcm for implantable mcdicalmicro-instruments. Journal of system Simulation. 2007; 19(15): 3582-35854