complex medical engineering - j. wu, et al., (springer, 2007)

J.L Wu, K Ito, S Tobimatsu, T Nishida, H Fukuyama (Eds.) Complex Medical Engineering J.L Wu, K Ito, S Tobimatsu, T Nishida, H Fukuyama (Eds.) Complex Medical Engineering With 274 Figures, Including in Color Springer Jing Long Wu, Ph.D., Professor Department of Intelligent Mechanical Systems, Faculty of Engineering, Kagawa University 2217-20 Hayashi, Takamatsu 761-0396, Japan Koji Ito, Dr.Eng., Professor Complex Systems Analysis, Adaptive & Learning Systems, Tokyo Institute of Technology 4259 Nagatsuta, Midori-ku, Yokohama 226-8503, Japan Shozo Tobimatsu, M.D., Professor and Chairman Department of Clinical Neurophysiology, Neurological Institute, Graduate School of Medical Science, Kyushu University 3-1-1 Maidashi, Higashi-ku, Fukuoka 812-8582, Japan Toyoaki Nishida, Dr.Eng., Professor Department of Intelligence Science and Technology, Graduate School of Informatics Kyoto University Yoshida Honmachi, Sakyo-ku, Kyoto 606-8501, Japan Hidenao Fukuyama, M.D., Ph.D., Professor Human Brain Research Center, Kyoto University Graduate School of Medicine 54 Shogoin Kawahara-cho, Sakyo-ku, Kyoto 606-8507, Japan ISBN-10 4-431-30961-6 Springer Tokyo Berlin Heidelberg New York ISBN-13 978-4-431-3096M Springer Tokyo Berlin Heidelberg New York Library of Congress Control Number: 2006930401 Printed on acid-free paper © Springer 2007 Printed in Japan This work is subject to copyright All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in other ways, and storage in data banks The use of registered names, trademarks, etc in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use Product liability: The publisher can give no guarantee for information about drug dosage and application thereof contained in this book In every individual case the respective user must check its accuracy by consulting other pharmaceutical literature Springer is a part of Springer Science+Business Media springer.com Typesetting: Camera-ready by the editors and authors Printing and binding: Asia Printing, Japan Preface In the twenty-first century, applications in medicine and engineering must acquire greater safety and flexibility if they are to yield better products at higher efficiency To this end, complex science and technology must be integrated in medicine and engineering Complex medical engineering (CME) is a new field comprising complex medical science and technology Included are biomedical robotics and biomechatronics, complex virtual technology in medicine, information and communication technology in medicine, complex technology in rehabilitation, cognitive neuroscience and technology, and complex bioinformatics This book is a collection of chapters from experts in academia, industry, and government research laboratories who have pioneered the ideas and technologies associated with CME Containing 54 research papers that were selected from 260 papers submitted to the First International Conference on Complex Medical Engineering (CME2005), the book offers a thorough introduction and a systematic overview of the new field The papers are organized into six parts Part focuses on biomedical robotics and biomechatronics and discusses principles and applications associated with the micropump, tactile sensor, underwater robot, laser surgery, and noninvasive monitoring Part discusses complex virtual technology in medicine, which involves visualization, simulators, displays, robotic systems, and walkingtraining systems In Part 3, the authors provide a comprehensive discussion of information and communication technology in medicine In Part 4, complex technology in rehabilitation is discussed, with topics including rehabilitation robotics and neurorehabilitation In Part 5, the authors discuss cognitive neuroscience and technology in five areas: complex medical imaging, including PET and MRI; human vision and technologies; brain science and cognitive technologies; transcranial magnetic stimulation (TMS); and electroencephalogram (EEG), neuron disease, and diagnostic technology In Part 6, the authors discuss topics associated with complex bioinformatics We first proposed the new term "complex medical engineering" for the First International Conference on Complex Medical Engineering (CME2005), which was successfully held in Takamatsu, Japan, in 2005 (http://frontier VI Preface eng.kagawa-u.ac.jp/CME2005/) When the conference was announced, we soon received a vast number of responses as well as support from the research community, industry, and many organizations To meet the strong demands for participation and the growing interest in CME, the Institute of Complex Medical Engineering (ICME) was founded in 2005 The ICME is an international academic society (http://frontier.eng.kagawa-u.ac.jp/ ICME/), the aim of which is to bring together researchers and practitioners from diverse fields related to complex medical science and technology ICME conferences are expected to stimulate future research and development of new theories, new approaches, and new tools to expand the growing field of CME The First Symposium on Complex Medical Engineering (http://frontier.eng.kagawa-u.ac.jp/SCME2006/) and The Second International Conference on Complex Medical Engineering (http://frontier.eng kagawa-u.ac.jp/CME2007/) will be held in Kyoto, Japan, and Beijing, China, respectively This book is recommended by the ICME as the first book on CME research It is a collaborative effort involving many leading researchers and practitioners who have contributed chapters on their areas of expertise Here, we would like to thank all authors and reviewers for their contributions We are very grateful to people who joined or supported the CME-related research activities, and in particular, to the ICME council members: Y Nishikawa, H Shibasaki, H Takeuchi, C.A Hunt, J Liu, J.C Rothwell, M Hashizume, M Hallett, M.C Lee, N Franceschini, N Zhong, P Wen, R Turner, S Miyauchi, S Tsumoto, W Nowinski, Y Deng, S Doi, T Touge, T Kochiyama, M Tanaka and R Lastra We thank them for their strong support Last but not least, we thank staffs at Springer Japan for their help in coordinating this monograph and for their editorial assistance Jing Long Wu Koji Ito Shozo Tobimatsu Toyoaki Nishida Hidenao Fukuyama Contents Preface V Biomedical Robotics and Biomechatronics Improving the Performance of a Traveling Wave Micropump for Fluid Transport in Micro Total Analysis Systems T SUZUKI, I KANNO, H HATA, H SHINTAKU, S KAWANO, and H KOTERA Vision-based Tactile Sensor for Endoscopy K TAKASHIMA, K YOSHINAKA, and K IKEUCHI 13 6-DOF Manipulator-based Novel Type of Support System for Biomedical Applications S Guo, Q WANG, and G SONG 25 The Development of a New Kind of Underwater Walking Robot W ZHANG, and S Guo 35 Mid-Infrared Robotic Laser Surgery System in Neurosurgery S OMORI, R NAKUMURA, Y MURAGAKI, I SAKUMA, K MIURA, M DOI, and H ISEKI 47 Development of a Compact Automatic Focusing System for a Neurosurgical Laser Instrument M NOGUCHI, E AOKI, E KOBAYASHI, S OMORI, Y MURAGAKI, H ISEKi, and I SAKUMA 57 Non-invasive Monitoring of Arterial Wall Impedance A SAKANE, K SHIBA, T TSUJI, N SAEKI, and M KAWAMOTO 67 VIII Contents Complex Virtual Technology in Medicine Advanced Volume Visualization for Interactive Extraction and Physics-based Modeling of Volume Data M NAKAO, T KURODA, and K MINATO 81 Development of a Prosthetic Walking Training System for Lower Extremity Amputees T WADA, S TANAKA, T TAKEUCHI, K IKUTA, and K TSUKAMOTO 93 Electrophysiological Heart Simulator Equipped with Sketchy 3-D Modeling R HARAGUCHI, T IGARASHI, S OWADA, T YAO, T NAMBA, T ASHIHARA, T IKEDA, and K NAKAZAWA Three-dimensional Display System of Individual Mandibular Movement M KOSEKI, A NllTSUMA, N INOU, and K MAKI 107 117 Robotic System for Less Invasive Abdominal Surgery I SAKUMA, T SUZUKI, E AOKI, E KOBAYASHI, H YAMASHITA, N HATA, T DOHI, K KONISHI, and M HASHIZUME 129 Information and Communication Technology in Medicine Attribute Selection Measures with Possibility and Their Application to Classifying MRSA from MSSA K HiRATA, M HARAO, M WADA, S OZAKI, S YOKOYAMA, and K MATSUOKA 143 A Virtual Schooling System for Hospitalized Children E HAN ADA, M MIYAMOTO, and K MORI YAM A 153 Using Computational Intelligence Methods in a Web-Based Drug Safety Information Community A.A GHAIBEH, M SASAKI, E.N DOOLIN, K SAKAMOTO, H CHUMAN, and A YAMAUCHI Analysis of Hospital Management Data using Generalized Linear Model Y TSUMOTO and S TSUMOTO 165 173 Contents IX Data Mining Approach on Clinical/Pharmaceutical Information accumulated in the Drug Safety Information Community A YAMAUCHI, K SAKAMOTO, and H CHUMAN Clinical Decision Support based on Mobile Telecommunication Systems S TSUMOTO, S HiRANO, and E HANADA 187 195 Web Intelligence Meets Immunology J LIU, N ZHONG, Y YAO, and J.L Wu 205 Complex Technology in Rehabilitation Hand Movement Compensation on Visual Target Tracking for Patients with Movement Disorders J IDE, T SUGI, M NAKAMURA, and H SHIBASAKI 217 Approach Motion Generation of the Self-Aided Manipulator for Bed-ridden Patients A HANAFUSA, H WASHIDA, J SASAKI, T FUWA, and Y SHIOTA 227 Lower-limb Joint Torque and Position Controls by Functional Electrical Stimulation (FES) K ITO, T SHIOYAMA, and T KONDO 239 Pattern Recognition of EEG Signals During Right and Left Motor Imagery ^ Learning Effects of the Subjects ~ K INOUE, D MORI, G PFURTSCHELLER, and K KUMAMARU 251 A Hierarchical Interaction in Musical Ensemble Performance: Analysis of 1-bar Rhythm and Respiration Rhythm T YAMAMOTO and Y MIYAKE 263 Comparison of the Reaction Time Measurement System for Evaluating Robot Assisted Activities T HASHIMOTO, K SUGAYA, T HAMADA, T AKAZAWA, Y KAGAWA, Y TAKAKURA, Y TAKAHASHI, S KUSANO, M NAGANUMA, and R KiMURA 275 X Contents Cognitive Neuroscience and Technology Influence of Interhemispheric Interactions on Paretic Hand Movement in Chronic Subcortical Stroke N MURASE, J DUQUE, R MAZZOCCHIO, and L.G COHEN 289 BOLD Contrast fMRI as a Tool for Imaging Neuroscience R TURNER 297 What can be Observed from Functional Neuroimaging? J RiERA 313 Human Brain Atlases in Education, Research and Clinical Applications W.L NOWINSKI 335 Deploying Chinese Visible Human Data on Anatomical Exploration: From Western Medicine to Chinese Acupuncture RA HENG, S.X ZHANG, Y.M XIE, TJ WONG, Y.R CHUI, and J.C.Y CHENG MEG Single-event Analysis: Networks for Normal Brain Function and Their Changes in Schizophrenia A.A lOANNIDES 351 361 MEG and Complex Systems G.R BARNES, M.I.G SIMPSON, A HILLEBRAND, A HADJIPAPAS, C WiTTON, and RL FURLONG 375 MEG Source Localization under Multiple Constraints: An Extended Bayesian Framework J MATTOUT, C PHILLIPS, R HENSON, and K FRISTON 383 Differential Contribution of Early Visual Areas to Perception of Contextual Effects: fMRI Studies Y EJIMA 397 Brain-machine Interface to Detect Real Dynamics of Neuronal Assemblies in the Working Brain Y SAKURAI 407 Development of a Simulator of Cardiac Function Estimation for before and after Left Ventricular Plasty Surgery Tatsushi Tokuyasu\ Akito Ichiya\ Tadashi Kitamura\Genichi Sakaguchi", and Masashi Komeda^ Kyushu Institute of Technology, Fukuoka, Japan "Cardiovascular Surgery, Kurashiki Central Hospital, Okayama, Japan ^Cardiovascular Surgery, Kyoto University, Kyoto, Japan Chapter Overview Cardiac function is evaluated with pressure-volume diagram, wall thickening function, and cardiac output These are approximately estimated by using echocardiography and cardiac catheterizations In order to make a surgical plan for left ventricular plasty, surgeons palpate the ventricular wall i.e cardiac muscle of a patient's heart, where the ventricular thickness, elasticity, and contractile stiffness of the left heart are diagnosed Cardiovascular surgeons are requested to have the skill to estimate postoperative cardiac function according to the surgical plan determined in the surgery We present a simulation system for cardiac function evaluation, where left ventricle models made of finite elements are built and estimate surgical efficiency for left ventricular plasty It is assumed that the ventricular wall of the models is composed of one-layer, elastic and isotropic material We employed a simple systemic circulation model, Windkessel, in order to compute the PV diagram of the left heart model This paper presents a heart modeling method based on finite elements, simulation algorism, and its online use of the FEM computation results Realistic trends of PV diagram are computed in the simulation Key Words Cardiac function, finite element method, and heart modeling 606 T Tokuyasu et al Introduction Recently, left ventricular plasty is conducted on patients' heart suffering from myocardial infarction and dilated cardiomyopathy These diseases significantly impair a pump function of heart, because the ventricular wall loses its thickness according to the progression of disease Laplace theory shows the stress of a ventricular wall increases depending on the radius in the minor axis of the ventricle in the same ventricular pressure, so that the surgeons remove the cardiac muscle around diseased area and sew up the removed area in the minor axis direction of the left ventricle We consider that an accurate estimate of postoperative cardiac function before surgery would yield surgical efficiency and enhance safety, so that we have developed an evaluation simulator iFor cardiac function [2] It is well known that finite element method (FEM) is one on the most effective method for heart analysis Therefore, we also employed FEM to simulate surgical efficiency evaluated from postoperative cardiac function Left ventricular models for before and after surgery are simply constructed, where cardiac muscular characteristics, e.g muscular fiber direction and cardiac muscular layer, are ignored Watanabe et al showed that one-layer and isotropic material of a finite element heart model not effect for PV diagrams In this paper, we present a heart modeling based on finite elements, a simulation method, and results Method 2.1 Finite element heart model Geometric data of a left ventricle is measured from the CT image of a patient suffering from a slight dilated cardiomyopathy The coauthor of cardiac surgeon judged that image does not impair to make a normal left heart model Firstly, we measured 800 three dimensional coordinates of the inner and outer wall of the left ventricle, and arranged them as the nodes of a finite element heart model Finally, 400 hexahedral elements compose a normal left heart model as shown in Fig The material property of the model is isotropic, elastic, and incompressible, and the Young's modulus of the elements provides contractile stiffness [1] of heart A homogenous Poisson's ratio of 0.4 is assigned on all elements We used a pre-post proc- Development of a simulator of cardiac function estimation for before and after left ventricular plasty surgery 607 essor MENTAT (MSC, Inc.)[3] to process mesh generation of the left heart model, and give it material characteristics Fig Left ventricular model based on finite elements 2.2 Pre-Post operative model Refer to the medical literature [1] and surgeon's experience, a preoperative model for myocardial infarction was built as shown in Fig 2(a) The wall thickness around infarcted area decreases, because cardiac muscle does not have any regenerative power, so that the ventricular wall of infarcted area is thinner than the other normal part The infarcted cardiac muscle loses its expansive and contractile properties In addition, most myocardial infarction patients have a tendency to suffer from high blood pressure and arterial sclerosis (a) Preoperative model (b) Post operative model Fig Finite element models before and after left ventricular plasty for myocardial infarction In Fig 2(a), the colored part indicates the thinned and stiff area due to myocardial infarction, where the wall thickness is 30 % thinner than the other normal wall and the constant Young's modulus of 150000 Pa is given through one cardiac cycle 608 T Tokuyasu et al The postoperative model is shown in Fig 2(b) It is built from the preoperative model, where the infarcted part is removed firstly and all nodes around the opened area are connected forcedly with the opposite nodes for closing the cut edge in the minor axis direction 2.3 Cardiac function evaluation Generally, a pressure-volume (PV) diagram, Emax, and stroke volume (SV) of left heart are used for evaluating cardiac contractile function This simulator also applies them to evaluate the cardiac function of the finite element models Fig shows a PV diagram, where the Emax shows the maximum contractile force of a heart and is derived by the gradient of the line which passes Vo and the end-systole point a-b b-c c-d d-a : Isovolumic contractile phase : Ejection phase • : Isovolumic relaxation phase : Filling phase 90 KM 110 120 130 140 Fig Pressure and volume diagram for one cardiac cycle and its explanation 2.4 Young's modulus We assumed that cardiac muscular property tends to nonlinear, so that updated Lagrange method is employed to analyze the heart models In this study, the Young's modulus of the model shows contractile force of cardiac muscle As mentioned above, a patient suffering from myocardial infarction tends to high blood pressure and arterial sclerosis, consequently we built a preoperative model as shown in Fig The preoperative model is composed of not only normal wall but also infarcted wall In order to identify the Young's modulus of the normal muscle for one cardiac muscle, a simple systemic circulation model, Windkessel, computed a PV diagram, as shown in Fig 4, corresponding with a patient affected with Development of a simulator of cardiac function estimation for before and after left ventricular plasty surgery 609 high blood pressure and arterial sclerosis, and Fig shows a thin wall model whose ventricular thickness is the same as the normal part of the preoperative model By using both of Fig and Fig 5, we determined the Young's modulus of the normal muscle for one cardiac cycle, where a relationship between the ventricular pressure and volume of the thin wall model traces the PV diagram of Fig The determined Young's modulus is shown in Fig dt M) an Fig PV diagram of a patient affected with high blood pressure and arteriosclerosis Fig Thin wall model of left ventricle —•—thfamcdre^on * infircled | -:.:.:.:.:.:^^ ••••^C 140000 120000 I 60000 'r'^V'"'':""""'• \ I ' "A 40000 20000 rnn.MMnm^ 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Fig Young's modules for cardiac cycle 610 T Tokuyasu et al 2.5 Simulation l\/lethod In our simulator, MSC.Marc (MSC, INC.)[4] computes a behavior of the heart models, and a Windkessel model as shown in Fig is linked to FEM in order to simulate the ventricular pressure and volume for one cardiac cycle of the heart models Fig shov/s an electrical circuit of Windkessel linked to the left heart model, where the differential equations for pressure, blood flow volume, and ventricular volume are derived and processed with Eular method The normal values of the cardiovascular parameters of Windkessel for a human weighing 60[kg] are shown in Table A simulation starts from the iso-volumic phase as shown in Fig Table Cardiovascular valve resistance and aorta compliance Rli : Rio : Rs : Rco : Rb : Csa : Rla : Rra : 0.001[mmHg/ml/sec] 0.01[mmHg/ml/sec] 1.0[mmHg/ml/sec] 35.2 [mmHg/ml/sec] 15.0[mmHg/ml/sec] 2.0[ml/mmHg] 6.3[mmHg] 3.0[mmHg] Parameters for Systemic Circulation System Vlv -.Volume of left ventricle [ml] Csa 'Compliance of systemic artery [mllmmHg] Qb 'Quantity to bronchial blood vessel [ml/secj Qco 'Quantity to coronary blood vessel [ml/secJ Qlv 'Quantity to left ventricle [ml/secj Qao 'Quantity to aorta [ml/secJ ORli 'Resistance of left ventricular inflow valve [mmHg/ml/sec] Rio 'Resistance of left ventricular outflow valve [mmHg/ml/sec] Rb 'Resistance of bronchial blood vessel [mmHg/ml/sec] Rco 'Resistance of coronary blood vessel [mmHg/ml/sec] Rs : Resistance of systemic artery [mmHg/ml/sec] Pla 'Pressure of left atrium [mmHg] Q Plv '.Pressure of left ventricle [mmHg] Pao :Pressure of aorta [mmHg] Atmospheric pressure Fig Electrical circuit of Windkessel model Development of a simulator of cardiac function estimation for before and after left ventricular plasty surgery 611 Pressure [mmHg] M The dilatation phase d-a is interpolated by the equation (9) V„ E Eui Volume [ml] Fig Simulation algorism for one cardiac cycle 2.5.1 Isovolumic systole Isovolumic systole starts at point a shown in Fig 8, where Etai shows the Young's modulus, Ptai is the left ventricular pressure, and VED is the enddiastolic volume At the next time increment, ta2='tai-\-dt(=0.Ol[stc]), Young modulus is increased to Eta2 in accordance with the cardiac cycle as shown in Fig 6, the pressure Ptai brings the difference of volume AVtai Pressure Pta2 which zeros the difference of volume AVtai is obtained and inputted to the Windkessel model in order to calculate Paotai- This process is repeated until Plv exceeds Pao, 2.5.2 Ejection phase The difference of the volume AVtti is calculated by the Windkessel model As Young modulus increases from Etbi to Etti, the pressure Ptti which corresponds to J K between the two volumes from Windkessel and the FEM model analysis is identified and inputted to Windkessel for simulating Paotbi' These processes are repeated until Pao exceeds Plv, 2.5.3 Isovolumic relaxation and diastole filling phase The same analysis as the one for isovolumic systole is applied to isovolumic relaxation phase until Plv lowers than Pla, Equation (1) interpolates diastole filling phase of the PV diagram, where Vsvm and Px are needed to pass the two points d {Ptdi and VES) and a {Ptai and VED) in Fig.2 612 T Tokuyasu et al Elastance[mmHg/ml] = PLy(t) (2) VLv(t)-Vo Since we focused on the contractile function evaluation of the left ventricle, the dilatation phase is interpolated with equation (1) [5], where VSVM is a parameter depending on the patient weight and Px is estimated by the ventricular shape and disease progress We assumed VSVM=150 and Px=12.9 for the normal heart Results Fig shows the preoperative model at the end-systole, where the left ventricle is expanded partially due to the blood pressure, because the infarcted part does not contractile function This phenomenon is also found in clinical cases The results of the postoperative model shown in Table show that both Emax and SV were improved The end diastole volume of the postoperative model approximately equals 103 [ml] These results meet the typical effects of ventricular plasty for myocardial infarction Table Simulation results for cardiac functional indices Cardiac function Thin wall Pre operative Post operative Vo[ml] 68.1 80.9 55.2 SV[ml] 27.0 15.5 21.1 Emax[mmHg/ml] 4.81 3.51 5.79 In order to compute one PV diagram by using the presented method, we needed about 20 hours including the time to build a finite element model and output the postoperative PV diagram It is necessary for training use of this simulator to shorten the computation time for the cardiac function of the postoperative models, so that we present a method for online output of the postoperative PV diagrams in the next chapter Development of a simulator of cardiac function estimation for before and after left ventricular plasty surgery Fig Simulation result of postoperative ventricle at end systole period Pre-operative model - • • • Thii waO model - L _ "^ S ; | Post operative model ; \ j 1 1 _, 80 i I I ) I f ] 100 r l 120 140 160 Vokim[m(| Fig 10 Simulation result for PV diagram before and after surgery la : Length of long axis ma: Length of minor axis Ip : Longitudinal position of the center of a surgical area Fig 11 Definition of a surgical region for online use of the simulator 613 614 T Tokuyasu et al ELv[nimHg/ml] ® Multi-spline interpolated point - One-dimensional spline interpolation line Etvcg E^y(tj) = interp elv(la, ma, lp)\ ^Lv(h ) ^ interp _ eh{la, ma, Ip); / ^ E^y(t„ ) = interp elv{la, ma, Ip); ^^ ^y(hi "O-o-O-OO-O Time[sec] Fig 12 Interpolation method for postoperative elastance curve based on a multi-dimensional interpolation technique correlation coefficient: 0.997 rq: f 0.1 0.2 0.3 -FEMrcsttW • Online oulpiil 0.4 OJ 0.6 0.7 0.8 Timelsec] Fig 13 Simulation results for interpolating post operative elastance curve Online Use For a quick output of the postoperative cardiac function, we employed a multi-dimensional spline function Firstly, a surgical area is defined by an ellipse which is derived with variables (lengths of long axis and minor axis) as shown in Fig 11 The parameter Ip shown in Fig 11 shows a center position of the ellipse on the left ventricle A PV diagram and equation (2) derive an elastance curve for one cardiac cycle We defined a function, interp_elv{la, ma, Ip}, based on the three dimensional interpolation function, where the elastance values ELv(t) at every 0.04sec are interpolated and outputted Finally, one dimensional interpolation makes their elastance data into an elastance curve for one cardiac cycle as shown in Fig 12 In this simulation, the calculation for interpolation function is computed by MATLAB Spline Tool Box Development of a simulator of cardiac function estimation for before and after left ventricular plasty surgery 615 Previous to the interpolation for the elastance curve, 24 kinds of the postoperative models where the combinations of the range of la\ 0~35mm, ma: 0~47mm, and Ip: 0~35mm were spesified were simulated to make the tensile product for the interp_elv{la, ma, Ip} This online output method enabled us to compute the postoperative cardiac function within about sec Fig 13 shows the online output result for elastance curve, where the surgical region was specified with la: 22mm, ma: 35mm, and Ip: 10mm The coefficient of correlation was 0.997 Discussion & Conclusion This study simplified the characteristics of the cardiac wall, so that twist behavior for heart contraction, fibrous direction, and excitation conducting system of left heart were ignored However, the simulation results could get good evaluation from the cardiac surgeon coauthors In Fig 7, the preoperative model shows typical cardiac function for myocardial infarction, where both the muscular contraction and the stroke volume decreased For building a preoperative model, we simplified the area of myocardial infarction into a set of one-layer elements with the constant Young's modulus However, in order to simulate a more realistic shape of the infarcted area with smooth change of thickness along the circumference, the diseased area of the preoperative model needs to be meshed in more detailed In order to save the calculation time of the simulator, we adopted dimensional interpolation method, where a surgical area was determined with variables This method enabled us to compute the postoperative cardiac function within about seconds We have to install a parameter that consider the models' dissymmetry, for example, a position of surgical area in the minor axis direction of the heart model, though we need huge computation time to make the four dimensional interpolation tensile The cardiovascular surgeons desire to research curative effects of the left ventricular plasty for dilated cardiomyopathy, because the pathogenesis of dilated cardiomyopathy is still unknown We will build a dilated cardiomyopathy model based on a patient's MRI data, and evaluate not only contractile function but also dilated function of left heart 616 T Tokuyasu et al Acknowledgment This work partly was supported by Research Fellowships of the Japan Society for the Promotion of Science for Young Scientists References Yutani Chikao (2002) Atlas of Cardiovascular Pathology T Tokuyasu, A Ichiya, T Kitamura, G Sakaguchi, and M Komeda (2004) A Simulation for Cardiac Function Evaluation before and after Left Ventricular Plasty, Proc of Mechanical Engineering Congress 69-70 (In Japanese) http://www.mscsoftware.co.jp/solutions/software/p_marc_m.htm http://ecust.isid.co.jp/public/product/marc/outline/index.html The Japan Society of Mechanical Engineers (1991) Bioengineering Division Key Word Index action potential 565 adaptive beam former 500 approach motion 227 arterial wall impedance biosensor assistant force 218 auditory cortex 535 awake background eeg 489 aware shakes of hand 25 68 beam former 375 behavioral task 407 bereitschaftspotential 439 bernstein's problem 573 bilateral integration 424 biosensor 555 blood oxygenation level dependent (BOLD) 297 brain 297 brain atlas 335 brain computer interface (BCI) 251 brain-machine interface 407 cardiac function 605 cell assembly 407 clinical decision support 195 clinical QSAR 187 cochlear implant 535 cognitive system 263 communication 263 compensation of hand movement computational intelligence 165 computer controlled system 47 218 conductance injection 565 contextual effect 397 cortical cell 565 cortical field potential 447 cortical neural functions 512 cross-modal integration 424 data mining 143 decision-making 447 decision tree 143 digital subtraction angiography drug discovery 165 drug safety 165 595 electroencephalogram, electroencephalography (EEG) 251, 313, 439, 489 E-health 195 electrophysiology 107 electro-vascular coupling 313 empirical bayes 383 endoscopy 13 event-related desynchronization 439, 500 event-related potential 545 expectation maximization 383 explicit memory 517 factor 375 finite element method fluid transport 605 618 Key Word Index functional magnetic resonance imaging (fMRI) 297, 397, 460 focusing 57 functional electrical stimulation (FES) 239 functional neurosurgery 335 hearing loss 535 heart modeling 605 hemodynamics 469 hierarchical models 383 hierarchical somatosensory processing 424 hospital management 173 hospitalized children 153 human brain mapping 335 human breast milk 187 hybrid support system 25 ill-posedness of inverse 573 image processing 57 immunology 205 immunosensor 555 implicit memory 517 independent component analysis 407 information-community system 187 intention 439 interactive extraction 81 interhemispheric inhibition 290 intraparietal region 424 inverse kinematics 573 inverse problem 313, 362, 383 J Japanese 460 joint torque 239 laparoscopic surgery 129 laser 57 lower extremity prosthesis 93 lower limb amputees 93 machine learning 143 magnetic field tomography (MFT) 362 magnetic stimulation 290 magnetoencephalography (MEG) 362, 375, 383, 500, 535 malignant brain tumors 47 manipulator 25 medical data mining 173 medical expert system 195 medical imaging 81, 117 medical robotics 35 medicinal products 187 micropump mid-infrared 47 minimally invasive surgery 25, 129 model-based compensator 218 model construction 489 model selection 383 modeling 107 monkey 447 motor imagery 251 motor recovery 290 motor unit potential (MUP) 527 movement disorders 218 multichannel surface EMG 527 multi-joint reaching 573 multimedia communication 153 multivariate analysis 173 music 263 mutual information 595 n N400 517 near intra-red spectroscopy 469 needle EMG 527 neural coding 565 neuroimaging 297 neuropsychological therapy 413 neuroradiology 335 neuroscience education 335 neurosurgery 57 nonlinear dynamic systems 313 normal aging 517 Key Word Index Optimization 25 Parkinson's disease 517, 545 patient-specific modeling 117 pattern recognition 251 peristimulus time histogram (PSTH) 527 pharmacoepidemiology 165 physics-based modeling 81 position control 239 positron emission tomography (PET) 535 pregnancy and teratogenic agents 187 prism adaptation 413 prom ax 375 random-dot motion 500 reaction time 290 reaction time measurement system 275 readiness potential 447 redundant robots 573 rehabilitation 413 rehabilitation robot 227 restricted maximum likelihood (ReML) 383 remote control 153 respiration 263 retinotopy 397 robot assisted activity 275 robotic surgery 47 schizophrenia 362 schooling 153 self-aided manipulator 227 SI 424 SII 424 simulation 107 single trial analysis 362 source localization 383 619 spatio-temporal interaction 397 speech perception 460 stroke 335 subcortical neural functions 512 subcortical stroke 290 superior temporal area 460 surgical robot 129 synchronized correlation 407 t tactile sensor 13 telemedicine 195 temporomandibular disorders 117 thin-plate spline 595 3D geometry 81 3D structure perception from motion 500 threshold 57 trans cranical magnetic stimulation (TMS) 512 underwater walking robot unilateral neglect 413 35 video camera 227 virtual anatomry 351 virtual reality 93 visible human research 351 visual cortex 397 visual target tracking test 218 visualization 351 volume representation 81 voluntary movement 439 walking training system web intelligence 205 X-ray CT 117 93 ... (l-V5o7 5'' )(a''-f )-( l-^So/S )(a- / ) (6) Substituting Eqs (3) and (6) into Eq (5), E is expressed as follows: E= 9kA ^-^ So/S ){a-f) \6y[^ {{ \-^ So'' I S'' ){a'' - f )-{ \-^ So I S ){a-f)f (7) 2.2 Methods...J.L Wu, K Ito, S Tobimatsu, T Nishida, H Fukuyama (Eds.) Complex Medical Engineering J.L Wu, K Ito, S Tobimatsu, T Nishida, H Fukuyama (Eds.) Complex Medical Engineering With... ISBN-10 4-4 3 1-3 096 1-6 Springer Tokyo Berlin Heidelberg New York ISBN-13 97 8-4 -4 3 1-3 096M Springer Tokyo Berlin Heidelberg New York Library of Congress Control Number: 2006930401 Printed on acid-free