Presenter: Yang-Min Huang Adviser: Dr. Ji-Jer Huang Chairman: Hung-Chi Yang 2013/4/10 Electrical Impedance Tomography :電阻抗斷層造影 1 Outline Introduction Paper review Motivations & Purposes Methods & Materials Result Future Works References 2 Introduction Electrical impedance tomography (EIT) • Injection current sources • Measurement voltages • Image reconstruction EIT:電阻抗斷層造影 3 Introduction Comparison of Imaging Techniques Imaging Imaging Cost($) Resolution(%) MRI Structural Functional Highest <0.1 X-ray CT Structural PET Ultrasound Advantages Disadvantages Soft-tissue, High resolution Expensive, Magnetic field Technique EIT limit Radiation, Difficult to distinguish the soft-tissue High <1 High resolution, Fast Functional Middle >3 Ration show the organs physiological function Low resolution, Radiation Structural Functional 1 Non-invasive, Fast Low resolution, High noise, Low Functional Lowest Bone reflect 1 Non-invasive, No radiation, Low resolution Portable MRI:核磁共振造影 X-ray CT:X光電腦斷層 PET:正子放射造影 Ultrasound:超音波 EIT:電阻抗斷層造影 4 Paper review(1) From:Do˘ga G¨ursoy*, Member, IEEE, Yasin Mamatjan, Andy Adler, and Hermann Scharfetter” Enhancing Impedance Imaging Through Multimodal Tomography” IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 58, NO. 11, NOVEMBER 2011 Purpose To investigate how much additional performance improvements can be expected by combining datasets of different modalities. EIT:電阻抗斷層造影 MIT:磁感應斷層造影 ICEIT:誘導電流電阻抗斷層造影 5 Paper review(1) Electrode configuration 6 Paper review(1) 7 Motivations & Purposes To get the real image for using FEM and Neural Network. To complete the algorithm for using Matlab. 8 Methods & Materials Poisson equation Algorithm The forward problem The inverse problem 9 Methods & Materials Poisson equation σ:導電係數 Ĵ : 電流密度 n:物體表面的法向量 10 Methods & Materials FEM for EIT forward problem Galerkin method Φ:voltage V:basis vector space σ:conductivity FEM:有限元素法 EIT:電阻抗斷層造影 Galerkin method:伽遼金方法 11 Methods & Materials Radial Basis Function(RBF) neural network RBF neural network :輻狀基底函數類神經網路 σ:變異數 SN :樣本總數 12 Methods & Materials Block diagram 13 電壓比較 電流比較 3 1.5 2 1 1 0.5 電流 電壓 0 0 -1 Result -2 -3 -4 0 5 10 15 20 25 節點編號 30 35 40 -0.5 -1 -1.5 26 45 28 30 32 34 節點編號 36 38 40 Verification Sample 4 2 1.8 3 1.6 2 1.4 1 1.2 0 1 0.8 -1 0.6 -2 0.4 -3 -4 -4 0.2 -3 -2 -1 0 1 2 3 4 0 14 42 Result Measured voltage for using different current, 15 train data Sample 4 Image Reconstruction 1000 900 3 Sample 4 1000 900 3 800 2 2 600 0 500 400 -1 1 600 0 500 400 -1 -4 -4 100 -3 -2 -1 0 1 2 3 -4 -4 4 -3 -2 Sample 1000 900 3 -1 0 1 2 3 2 1000 900 3 600 0 500 400 -1 1 600 0 500 400 -1 100 1 2 3 4 -3 -800 -3 -2 -1 0 1 2 3 4 -3 100 -2 -1 0 1 -200 -400 -600 -3 -1000 -4 -4 1000 4 -800 -3 -2 -1 2 3 4 0 1 2 3 4 900 3 1000 900 3 800 800 2 700 1 600 0 500 400 -1 700 1 600 0 500 400 -1 300 300 -2 200 -3 -4 -4 -1000 Sample 4 -2 -3 0 -1 -2 200 -4 -4 200 0 -600 300 200 -3 400 1 -400 2 -2 0 -200 -1 700 300 -2 -1 0 800 2 1 600 Sample 4 700 -2 0 -4 -4 4 800 -3 200 Image Reconstruction 4 -4 -4 1 -2 100 800 2 200 -3 1000 3 400 300 -2 4 600 2 200 -3 800 3 700 300 -2 1000 800 700 1 Image Reconstruction 4 100 -3 -2 -1 0 1 2 3 4 200 -3 -4 -4 100 -3 -2 -1 0 1 2 3 4 15 Future Works Paper review To simulate more samples of image pattern To improve the RBF neural network To complete the user interface 16 References P. Wang, H. Li, L. Xie, Y. Sun, “The Implementation of FEM and RBF Neural Network in EIT”, Proceedings of the 2009 Second International Conference on Intelligent Networks and Intelligent Systems, pp. 66-69, IEEE Computer Society, 2009. Do˘ga G¨ursoy*, Member, IEEE, Yasin Mamatjan, Andy Adler, and Hermann Scharfetter” Enhancing Impedance Imaging Through Multimodal Tomography” IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 58, NO. 11, NOVEMBER 2011 Ybarra, G. A., Q. H. Liu, G. Ye, K. H. Lim, R. George, and W. T. Joines, "Breast imaging using electrical impedance tomography (EIT)," Emerging Technologies in Breast Imaging and Mammography, Ed.: J. Suri, R. M. Rangayyan, and S. Laxminarayan, American Scientific Publishers, 2008. 黃俊惟,電阻抗斷層成像技術之研究,南台科技大學電機工程研究所碩士 論文,2010 17