' '
Proceedings of the 2004 lntemationai Conferonco on intelligent Mechatronics and Automation Chengdu,Chlna August 2004
Robotic Neurosurgery and Clinical Applications Weimin Shen and Jason'Gu Department of Electrical & Computer Engineering Dalhousie University Halifkc, Nova Scotia B3J 2x4,Canada {~ 5 0 3 5 0 & 7 Jason.Gu}@dal.ca
-
Abstract This paper presents robotic neurosurgery and clinical applications. Firstly, background, objective, motivation and distinct properties of this kind of surgical robots are introduced. Secondly, by analysing current main classifications for surgical robots, neurosurgical robots are divided into two groups, namely, neurosurgical robots anid tele-neurosurgical robots. Then, after the history of surgical robots is briefly reviewed, several prototypes of neurosurgical robots currently used in clinics are discussed in details. As a branch of surgical robots, the design issues about neurosurgical robots not only have the common part as surgical robots, but this field has its special issues, such as high complex image modeling and analysis, high accuracy, and high safety. Therefore, current research challenges are provided and consideratialn about tele-robotic neurosurgical design is put forward. Finally, this paper is concluded with future directions.
-
Index Terms Tele-robotics. Surgical' robotics Time-delay control. S a f w issue. Tele-neurosurgety.
I. INTRODUCTION Head injury is a significant cause of morbidity and mortality in the injured population. Toda:y, trauma is the third most important cause of death overall, followed by cardiovascular disease and cancer, and remains the number one killer of human beings under the age of 45 in the western countries. In Canada, there are over 9,000 injury cases included with an Injury Severity Score (ISS) > 12, and over 1,200 deaths every year [ 11. The most common type of trauma is a head injury (5 1%). In addition, head injury itself remains the number one killer in major trauma and places an enormous burden on families, communities and the health care system. Economic, populous and geographica1 situation does not allow for every person to have access tal appropriate medical treatment. Because head injury is a challenging clinical entity for most doctors working in the province's emergency departments, most patients have to be bansferred to primary trauma centers. But the fact is that with head injure the response timq should be within one hour. This is not possible in certain areas even with use of the EHS air transport. Even for chronic conditions the need for a method of providing treatment in a timely and cost effective manner is not currently possible in many areas (remote communities, mining, lumbering, offshore). As a result, some ]people die as a result of the traumatic event prior to reaching acute care. Fortunately, by combining the advantageous outcomes of medical and robotic fields, we can firid an answer to this challenge'issue. For example, the remota drilling of a hole in the skull of human beings to relieve the pressure associated
0-7803-8748-1/04/$20.00 02004 mEE.
Evangelos Milios Faculty of Computer Science Dalhousie University Halifax, Nova Scotia B3H I W5,Canada
[email protected]
with head trauma and chronic conditions, will enhance initial management of patients with head injure, saving time and money in transport and providing the patient with their immediate family and support group. Obviously, by robotic neurosurgery system, patients, surgeons and societies may experience such main benefits as helping remote patients get acute care, leading to improved patient outcome, decreasing mortality in the injured population, reducing health care cost, and so on. As a matter of fact, neurosurgery is very complex, and it needs high expert surgeons with strong experience. In fact, neurosurgery was the first clinical application of robotics and continues to be a topic of current interest. Recently, after tele-robotics was applied to medical application, tele-robotics neurosurgery system became more and more important and popular. Therefore, this paper will give a detail review about robotic neurosurgery and clinical applications. The rest of the paper is organized as follows. Section 2 briefly discusses the classification for neurosurgical robots. Section 3 describes the historical review of surgical robotics first, then gives typical neurosurgery applications, finally puts forward hture development of robotic neuro-surgery. The challenging issues in tele-robotic neuro-surgery are discussed in section 4. The conclusion is given in Section 5. 11. CLASSIFICATION OF NEUROSURGICAL ROBOTS There are many different classification strategies for surgical robots, by manipulator design (e.g., kinematics, actuation), by level of autonomy (e.g., reprogrammed versus teleoperation versus constrained cooperative control), by targeted anatomy or technique (e.g., cardiac, intravascular, percutaneous, laparoscopic, micro-surgical), intended operating environment [e.g., in-scanner, conventional operating room (OR)], etc [2]. In this section, several main classification strategies in surgical robotics will first be introduced, then, it will give one classificationstrategy to classify neurosurgericalrobots. The robots are classified by the role they play in medical applications, which was given by Taylor [3]. He stresses the role of robots as tools that can work cooperatively with physicians to carry out surgical interventions and identifies five classes of systems: Intern replacements Tele-surgical systems Navigational aids Precise positioning systems Precise path systems .; I
114
MAIN 1
I
System
I
IGOR
I I
Institution
Grenoble Univ.
Minerva
Univ. of Lsusanne
KEN-MRI
Univ. of Tokyo Integrated Surgical
.^
1 I..:..
USA
1985
Guidance /Control CT Guided
Germany
1989
MRI Guided
Country
Year
France
1992 Preoperat. Imaging, Tracking
Switzerland 1993
Human
Table
Positioning Arm Puma 200
Human
Table
6 W F ARM
Applied to
Mount
Human
Floor
X-Ray Guided
Human
Floor
MRI Guided
Japan
1995
Tissue samples
Table
USA
1996 Preoperat. Imaging, Tracking
Human
Floor Trolley
Japan
1996
Human
Floor
1997 Preoperat. Imaging, Tracking
Human
Floor
China U.K.
’
2000 Preoperat. Imaging, Tracking
Singapore 2002 Preoperat. Imaging, Tracking China
2003 Preoperat. Imaging, Tracking1
Passive, semi-active, and active robotic systems are as three classes of surgical systems described by Troccaz [4]. In Dario’s paper [ 171, surgical robots are classified as: Large, high-precision robots Hand-held smart medical tools 0 Miniature invasive robots In Taylor’s recent paper, he has chosen to focus on the role of medical robots within the context of their role in CIS systems [2]. He classifies the systems into two broad families: surgical CAD/CAM and surgical assistants. In Kevin’s paper [181, by clinical application, surgical robots are divided, since clinical applications are more interesting to the end-user. Main neurosurgical robots are listed in table 1. They can be simply classified into two families: neurosurgical robots and tele-neurosurgicalrobots. 111. STATE O F ART IN NEUROSURGICAL ROBOTICS
Neurosurgical robotics is one branch of surgical robotics, and the first surgical robot was developed just for neurosurgical application. Therefore, first, shortly historical review of surgical robotics will be given as followings, next, typical robotic neurosurgical systems will be introduced, finally, robotic neurosurgery future development will be discussed. A . Historical review of surgical robotics Robotics was first applied to the industry field. Later, it was introduced into the medical field. In fact, surgical robotics is a relatively young field, with the first recorded medical appiication of a robot occurring in 1985 IS]. In this case, the robot was a simple positioning device to orient a needle for biopsy of the brain. In 1985, a 52-year-old man was put on a CT scanner couch. CT pictures of the brain were taken, the target was identified on the CT pictures, the robot was brought in operation and quickly the robot oriented the bushing towards the suspected area in the brain. The, biopsy of tissue was sucked with a syringe. The issue sample was sent to the
3 Trans1 DOF
PUMA 260
Floor Human Human
Table
I
Floor
7 DOF paralle manipulator
6 DOF
pathology lab, where a positive biopsy was confirmed on the first sample. Shortly thereafter, research groups in Europe, Asia, and the United States began investigating medical applications of robotics. In Europe, a group at Imperial College in London under the direction of Davies began developing a robot for prostate applications [ 191. At Grenoble University Hospital in France, Benabid, Lavallee, and colleagues started work on neurosurgical applications such as biopsy [20]. In Asia, Dohi at Tokyo University developed a prototype of a CT-guided needle insertion manipulator [21]. In the U.S., Taylor and associates at IBM began developing the system later known as ROBODOC [22]. Currently, there are several commercial ventures and a handful of research laboratories active in the field of medical robotics. These early research efforts have led to some commercial products, for example, NeuroMate robot of Integrated Surgical Systems [l 11, Da Vinci surgical robot system [23], Zeus tele-surgical system [24], and so on.
-
B. Robotic neurosurgery applications As a matter of fact, neurosurgery was the first clinical application of robotics and continues to be a topic of current interest. In general, neurosurgery is very complex, so that it needs more expert surgeons and higher accuracy. However, robots have accuracy position system and high stiffhess so that they can eliminate the quivering of the surgeon’s hand resulting in a more delicate surgery. Then, they reduce trauma during surgery. One of the earliest robotic systems developed for precise needle placement was the neurosurgical robot Minerva [8, 91, designed for stereotactic brain biopsy. A special purpose robot was constructed which was designed to work within the CT scanner so that the surgeon could follow the position of the instruments on successive CT scans. This constraint ensured that CT images would be available throughout a procedure, keeping all procedures under the surgeon’s supervision and 115
control. A diagram of the system and associated components is shown in Fig 1. The system consists of a five degree of freedom structure with two linear axes (vertical ana lateral), two rotary axes (moving in a horizontal and vertical plane), and a linear axis (to move the tool to and from the patient's head). The robot is mounted on a horizontal carrier, which moves on rails. A stereotactic frame, ihe Brown-RobertsWells ,(BRW) reference frame, is attached to the robot gantry and coupled to the motorized CT table by two ball and socket joints arranged in series. The system was used for two operations on patients in September 1993 at the CHUV Hospital in Switzerland, but the project has since been discontinued.
can be digitized (radiographs, for example) using a digitizing table or scanner. Once the patn is planned, the images are transferred directly from the planning workstation to the control workstation in the operating room over an Ethernet link.
Fig. 2 NeuroMate neurosurgical robot
Fig. 1 Minerva componentsand system overview Fig. 3 Calibration cage held by the robot
The NeuroMate is a six-axis robot for neurosurgical applications that evolved from work done by Benabid, Lavallee, and colleagues at Grenoble University Hospital in France [6, 20, 251. The original system was subsequently redesigned to fulfill specific stereotactic requirements and particular attention was paid to safety issues [26]. The current version, shown in Fig. 2, is a commercial product that has been licensed by Integrated Surgical Systems (Davis, California, USA) and is FDA approved. The system has been used in over 1600 procedures since 1989, covering a range of neurosurgical procedures. The major clinical applications include: tumor biopsies (1 100 cases), stereoelectroencephalographic investigations of patients with epilepsy (200 cases), midline stereotactic neurosurgery and functional neurosurgery of the basal ganglia (200 cases). A typical clinical procedure consists of an initial data acquisition step, followed by data transfer to the control computer, and then the procedure itself. Data acquisition involves obtaining images of the brain from which path planning from the skin entry point to the target point can be done using a specially developed sofware program. The images can be in digital form (DSA, CT,or MRI images) or
To carry out the procedure, the robot must know where it is located relative to the patient's anatomy. This is typically done using a calibration cage, which is placed on the endeffector of the robot around the patient's head shown in Fig. 3. This cage looks like an open cubic box and the four sides are each implanted with nine X-ray opaque beads, the positions of which have been precisely measured. Two X-rays are taken which show the position of these beads along with the fiducial markers of the patient's frame. This information is used to determine the transformation matrix between the robot and the patient. The defined trajectory is used to command the robot to position a mechanical guide, which is aligned with this trajectory. The robot is then fixed in this position and the physician uses this guide to introduce the surgical tool such as a drill, probe, or electrode. At the first stage, neurosurgerical robots have been developed only used CT images for guidance. Later, many structures in the brain are best visualized using magnetic resonance imaging (MRI). The robotic systems described so far are not suitable for use in an MRI scanner because the
116
~
strong magnetic fields generated dictate that only nonmagnetic materials can be used. In Japan, in the Mechatronics Laboratory at the University of Tokyo, Dohi, Masamune and colleagues developed an MRI-compatible needle insertion manipulator intended for use in stereotactic neurosurgery [101. The manipulator frame was manufactured using polyethylene terephthalate (PET) and ultrasonic motors were used for the actuators. Other parts such as bearings, feed screws, and gear that must be strong and precisely fabricated are made of nonmagnetic materials including brass, aluminum, delrin, and ceramics. In phantom tests using watermelons, the robot performed satisfactorily with a positioning error of less than 3.3 mm from the desired target. The unit was small enough at 491mm in maximum height to fit inside the MRI gantry of 600 mm in diameter. Fig. 5 MRI compatible robot in interventional MRI system
Fig. 4 MRI compatible robot design
Fig. 6 Robot-assisted neurosurgery system
Rather than retrofitting an industrial robot, Masamune developed a completely new design based on the clinical requirements for safety, MFU compatibility, and compactness. As shown in Fig. 4, the system includes an X - Y - Z base stage. An arch mechanism is mounted on the base stage along with a linear needle carriage. This isocentric design was adopted for its mechanical safety and simplicity. The system was controlled by a personal computer. The control computer and motor driver boards were remotely located in the MRI control room and connected by shielded cables to the robot. In a related development, a new MRI compatible robot has been developed to work within the interventional MRI unit at the Brigham and Women's Hospitat in Boston, Massachusetts, USA [27]. The interventional MRI has a pair of parallel facing donut-shaped magnets, with an air gap of 560 mm. The robot sits between the magnets and is mounted on at the top of the unit as shown in Fig. 5. The system is currently undergoing testing, and one potential clinical application is needle placement for prostate brachytherapy. Finally, researchers in Germany have developed an MRI compatible robotic biopsy system, focusing on breast cancer as an initial application. In vitro experiments using pig livers in a 1.5 Tesla magnet and 4 mm targets resulted in all eight targets being successfidly hit [28].
117
Since 1995, a research on robot assisted minimally invasive neurosurgery was done in China, aiming at development of a prototype system for this kind of intervention of robot-assisted. neurosurgery, shown in Fig. 6 [13, 16,291. Recently, they succeeded to apply tele-robotics to neurosurgery. This system consists of three components: image-guided surgical planning and supporting system, marker-based registration with some measurement tools, a 6 degree-of freedom PUMA 260 robot execution with customdesigned interface between the end-effector and surgical tool. The purpose of this system is to make planning the trajectory of probe on the graphic computer, to locate the target of tumor in the brain, to avoid the exposure of radiotherapy during injecting dangerous radio isotope via remote control, with collaboration of engineering. NEUROBOT is a robotic neurosurgery system which is being built in Nanyang Technological University in a collaborative project with National Neurological Institute (Singapore) for the skull base surgeries [15, 301. As shown in Fig. 7, this is an image-guided surgery system which consists mainly of three modules: 1) an image-guided motion planner & controller, 2) a parallel robot, and 3) an optical position tracking system. In the planning phase, both 2-D image and 3D model of patient's skull are presented to the surgeon for
improve surgical training worldwide. Some robotic surgeries better topographic assessment of the tumor and the skull-base anatomy. No-go areas will be marked by the surgeon. Once have been used by tele-operation, and more and more telerobotic neurosurgery will be employed in the future. the surgeon has defined the areas that an: to be avoided, the motion planner automatically proceed:; to generate the 3) Friendly Human-Machine Interface: All of the required drill path using a proprietary path-generation common interfaces (mice, joysticks, touch screens, push algorithm. The parallel manipulator carties tool-holder to buttons, foot switches, etc.) used for interactive computer perform accurate bone-remove tasks. The manipulator is applications are used to provide input for surgical systems as functionally divided into three layers: the base suppott, M-850 well. However, surgeons routinely use voice to communicate Hexapod, and a 7th DOF mounted on the !Hexapod. Following -with operating room personnel. Further, their hands (and feet) are frequently rather busy. Accordingly, there has long been the path generated by the motion planner, the manipulator interest in using voice as a two-way command and control performs the task of bone removal while tlhe surgeon monitors system for surgical applications. the entire process. The cavity generated by the robot allows access to deep seated brain areas, which are inaccessible by Iv. THECHALLENGING ISSUESIN TELE-ROBOTIC other routes. It is expected that this system will help neuroNEUROSURGERY and ENT surgeons to reduce skull drillirig time (from 5 - 8 hours to less than 2 hours). The accuracy required for such Tele-robotic neurosurgery is the tendency of next operations is 0.5mm. The optical tracking system, generation of robotic neurosurgery. As a branch of robotic OPTOTRACK 3020, consists of infrared light-emitting diode surgery, not only does tele-robotic neurosurgery have the same markers (IREDs) and lateral-effect photodiode cameras. The technical problems in the field of robotic surgery, but also it cameras track the positions of IRED markers placed on a rigid has its special problems as follows. body to determine the object's position and orientation in its I) Imaging, modeling, and analysis issue: Advances are measuring frame. The measurement system has an accuracy of needed in techniques for building patient-specific anatomical 0.1" and resolution of 0.01". By simultaneously tracking models from preoperative images and real-time sensor data, more than 256 markers, the system can detect displacements for incorporating biomechanical information into these in 6 DOFs in complex applications. In this work, the tracking models, and for using this information to help control the system is used for the position and orientation measurement. robot. Similarly, we need much better ways of modeling surgical procedures and of using this information both in pretreatment planning and real-time execution. 2) Accuracy issue: The brain is the powerhouse of the body, even though it only makes up two percent of the body's weight. For skull-base surgery, the robot deals with one of the most complex regions of the human anatomy. Vital blood vessels and major cranial nerves pass through the skull base. Two carotid arteries, two vertebral arteries and twelve pairs of cranial nerves reside in that region along with the blood drainage system of the brain. Any damage to them will cause serious problem in patient's recovery or, even worse, to lose some functions. In general, requirement for accuracy is less than 0.1". To solve this issue can use parallel architecture robot or employ high-accuracy calibration system. 3) Time delay control issue: Tele-operation will confront Fig. 7 Scheme of NEURobot System time delay because of communication via the internet. The C. Future Development of Robotic Neurosurgery limitation of the acceptable time in terms of surgeon's I ) Special-purpose Architecture: The first generation of perception of safety is 330ms. At the same time, high speed surgical robotic systems in the 1980s employed generaland real time communication is indispensable especially for purpose industrial manipulators, either directly or with minor the neurosurgical application to realize reliable diagnosis and modifications ([5], [31]). Industrial robots are still being used appropriate treatment. To decrease the time delay can use high today as research and validation tool:; where immediate speedwide band optical fiber network and ATM. clinical use is not contemplated or specialized kinematic 4) Safety issue: Safety is extremely important for robotic design is not essential. Such systems are robust, available, and neurosurgery. To enhance the safety of medical robot, often have open interfaces suitable for iexperimentation [2]. emergency button, redundancy sensor, and mechanical brake However, special-purpose architecture for robots becomes were adopted. One such approach is the Hazard Identification more and more desirable, and it will be the tendency of next and safety Insurance Control (HISIC) [32], which attempts to generation of robotic neurosurgery. incorporate the possibility of failure in software and hardware 2) Tele-operated The development of communication of the system independently, and then together. By technology makes it possible to project surgical expertise into considering the seriousness of any fault, and acceptability of occurring, a final safety index for the whole system is given. remote locations for difficult or rare operations, and to
118
V. CONCLUS~ON ANDFUTURE DIRECTION
When the first surgical robot was applied in 1985, it was dificult to be accepted by people. But robotic surgery can make complicated procedures much safer, more effective. In particular, robotic neurosurgery, can make many procedures dramatically safer and less expensive, and can bring great benefit to both surgeons and patients, so that it becomes more and more common. In the future, the reseach about neurosurgial robotics will have the following tendency. First, neurosurgical robotics will employ special-purpose architecture instead of generalpurpose architecture, which will help improve performance of robots, particularly in safety issue. Second, neurosurgical robotics will use tele-operation to open its wide applications, which will provide remote patients with access to high-level health care. Finally, as a kind of friendly human machine interface, voice will be widely employed in neurosurgical robots; which can make neurosurgical robots more practical. With the development of technolo&, tele-robotic neurosurgery will become a promising field, and it can. meet safety need very well, be easier to control, let patient recover more quickly. REFERENCES
[ I ] National Trauma Registry, 2003 Report: Major Injury in Canada, 2003, pp. 7-15. [2] R. H. Taylor, D.Stoianovici, “Medical Robotics in Computer-Integrated Surgery”, IEEE Transactions on Robotics and Automation, vol. 19, pp. 765-781, October 2003. [3] R. H. Taylor, “Robots as. Surgical Assistants:, Where We Are, What We Are Tending, and How to Get There”, in Proceeding of the 6Ih Conference on Artificial Intelligence in Medicine in Europe (AIME 97), Grenoble, France, March 1997, pp. 3-1 I . [4] J. Troccaz and Y. Delnondedieu, “Robots in Surgery”, IARP Workshop on Medical Robots, Vienna, Austria, 1996. [5] Y. S. Kwoh, J. Hou, and E. A. Jonckheere et al., “A Robot with Improved Absolute Positioning Accuracy for CT-guided Stereotactic Brain Surgery,” IEEE Transactions on Biomedical Engineering, vol. 35, pp. 153-161, February 1988. [6] S . Lavallee, J. Troccaz, et al., “Image-guided Operating Robot: a Clinical Application in Stereotactic Neurosurgery”, in ComputerIntegrated Surgery, R. H. Taylor, S . Lavallee, G . C. Burdea, and R. Mosges, Eds. Cambridge, MA: MIT Press, 1996, pp. 343-352. [7] P. Cinquin, J. Troccaz, et al., “IGOR: Image Guided Operating Robot,” Innovation at Technological Biological Medicine, vol. 13, pp. 374-394, 1992. [SI D. Glauser, P. Flury, N. Villotte, and C. Burckhardt, “Mechanical Concept of the Neurosurgical Robot Minerva”, Robolica, vol. 1 1, pp. 567-575, 1993. [9] C. W. Burckhart, P. Flury, and D. Glauser, “Stereotactic Brain Surgery”, IEEE Engineering in Medicine and Biology, vol. 14, pp. 3 14-3 17, 1995. [IO] K. Masamune, E. Kobayashi, et al., “Development of an MRIcompatible Needle Insertion Manipulator for Stereotactic Neurosurgery”, Joumal of Image Guided Surgery, vol. I , pp. 242-248, 1995. [If] S. Lavallee, J. Troccaz, et al., “Image-guided Operating Robot: a Clinical Application in Stereotactic Neurosurgery”, in ComputerIntegrated Surgery, R. H. Taylor, S . Lavallee, G. C. Burdea, and R. Mosges, Eds. MIT Press, 1995, pp. 343-351. [12] F. Arai, M. Tanimoto, et al., “Multimedia Tele-surgery Using High Speed Optical Fiber Network and Its Application to Intravascular Neurosurgery”, Proceeding Of the 1996 IEEE Intemational Conference on Robotics and Automation, Minneapolis, Minnesota, April 1996, pp. 878-883.
119
[13] M. D. Chen, T. Wang, et al., “A Robotics System for Stereotactic Neurosurgery and Its Clinical Application”, Proceeding Of the I?98 hternational Conference on Robotics & Automation, Leuven, Belgium, May 1998, pp. 995-1000. B. Davies, S . Starkie, et al., “Neurobot: a Special-purpose Robot for Neurosurgery”, Proceeding Of the 2000 IEEE International Conference on Robotics and Automation, San Francisco, CA, April 2000, pp. 41034108. [15] S. P. Bai, M. Y. Teo, “A Robotic Neuro-Surgery System and Its Calibration by Using a Motion Tracking System”, Proceeding Of the 2002 IEEE International Workshop on Robot and Human Interactive Communication, Berlin, Germany, September 2002, pp. 436-441. [16] W. S . Chou, T. M. Wang, D. Liu, Z. M. Tian, “Computer and Robot Assisted Tele-neurosurgery”, Proceeding Of the 2003 IEEEIRSJ International Conference on Intelligent Robots & Systems, Las Vegas, Nevada, October 1998, pp. 3367-3372. [I71 P. Dario, A. Menciassi, “Robotics for Surgery”, Proceeding Second Joint EMBS/BMES Conference, Houston, T X , USA, 2002, pp. 18131814. [18] Kevin Cleary, Charles Nguyen, “State of the Art in Surgical Robotics: Clinical Applications and Technology Challenges”, Computer Aided Surgery, vo1.6, pp.312-328,2001. [I91 B. L. Davies, R. D. Hibberd, e/ al., “A Surgeon Robot for Prostatectomies“, presented on the 5“ International Conference ’ on Advanced Robotics (ICAR91). 1991. [20] A. L. Benabid, P. Cinquin, et al., “Computer-driven Robot for Stereotactic Surgery Connected to CT Scan and Magnetic Resonance Imaging Technological Design and Preliminary Results”, Application Neurophysiology, vol. 50, pp. 153-154, 1987. [21] Y. Yamauchi, T. Dohi, et al., “A Needle lnsertion Manipulator for X-ray CT Image-guided Neurosurgery”, Proceeding of LST, vol. 5, 1993, pp. 8 14-821. [22] R. H. Taylor, B. D. Mittelstadt, et al., “An Image-directed Robotic System for Precise Orthopaedic Surgery”, IEEE Transactions on Robotics and Automation, vol. 10, pp. 261-273, 1994. [23] G. S . Guthart and J. J. Kenneth Salisbury, “The Intuitive Telesurgery System: Overview and Application”, IEEE International Conference on Robotics and Automation, 2000, pp. 618-621. [24] D. H. Boehm, 8.Reichenspumer, et al., “Clinical Use of a Computerenhanced Surgical Robotic System for Endoscopic Coronary Artery Bypass Grafting on the Beating Heart [In Process Citation]”, Thorac Cardiovasc Surgery, vol. 48, pp. 198-202,2000. [25] S . Lavallee, “A New System for Computer Assisted Neurosurgery”, Proceedings of the Eleventh IEEE Engineering in Medicine and Biology Conference, 1989, pp. 926-927. [26] A. L. Benabid, D. Hoffmann, et al., “Robotic Guidance in Advanced Imaging Environments”, in Advanced Neurosurgical Navigation, E. A. 111 and R. J. Maciunas, Eds. New York: Thieme Medical Publishers, Inc., 1999, pp. 571-583. [27] K. Chinzei, N. Hata, et al., “MR Compatible Surgical Assist Robot: System Integration and Preliminary Feasibility Study”, Medical Image Computing and Computer Assisted Intervention (MICCAI), pp. 921 -930, Springer, 2000. [28] W. A. Kaiser, H. Fischer, et al., “Robotic System for Biopsy and Therapy of Breast Lesions in a High-field Whole-body Magnetic Resonance Tomography Unit”, Invest Radio, vol. 35, pp. 5 13-519,2000. [29] D. Liu, T. M. Wang, et al., “Study on Robot-Assisted Minimally Invasive Neurosurgery”, Proceeding of the 2001 IEEE International Conference on Robotics & Automation, Seoul, Korea, May 2001, pp. 2008-2013. [30] C. Sim, W. S . Ng, et al., “Image-guided Manipulator Compliant Surgical Planning Methodology for Robotic Skull-base Surgery”, International Workshop on Medical Imaging and Augmented Realip (MIAR ‘Ol), Shatin, N. T., Hong Kong, June 2001, pp. 26-29. [31] J. L. Garbini, R. G. Kaiura, et al., “Robotic Instrumentation in Total Knee Arthroplasty,” in Proceeding 33rd Annual Meeting, Orthopaedic Research Society, San Francisco, CA, 1987, pp. 413. [32] Baowei Fei, Wan Sing Ng, Chee Keong Kwoh, “The Hazard Identification & Safety Insurance Control (HISIC) for Medical Robot”, Proceedings of the 22nd Annual EMBS International conference, Chicago IL,July 2000, pp. 3022-3026.
