(Tel: ; )

Transcription

1 th International Conference on Control, Automation and Systems (ICCAS 2013) Oct , 2013 in Kimdaejung Convention Center, Gwangju, Korea Prototype Modular Capsule Robots for Capsule Endoscopies Laehyun Kim l * 2, Sai Chun Tang, and Seung-Schik 3 Yoo i Center for Bionics, Korea Institute of Science and Technology, Seoul, , Korea (Tel: ; laehyunk@kist.re.kr) * Corresponding author 2 Department of Radiology, Harvard Medical School / Brigham and Women's Hospital, Boston, 02115, USA (Tel: ; sct@bwh.harvard.edu ) 3 Department of Radiology, Harvard Medical School / Brigham and Women's Hospital, Boston, 02115, USA (Tel: ; yoo@bwh.harvard.edu ) Abstract: Capsule endoscopy allows clinicians to wirelessly examine the small intestine using a capsule equipped with a miniscule camera. However, it has intrinsic limitations such as a lack of controlled capsule locomotion and limited therapeutic functions inside the gastrointestinal (GI) track. Recently, several researches have been conducted to prototype robotic capsules that have a self-propelling mechanism by integrating modern technologies. To routinely use in a clinical setting, several technical challenges, including size constraint of the capsule, locomotion mechanism, and stable power source, should be addressed. In this paper, we introduce a prototype of a modular robotic capsule system, which is designed to distribute the functional burdens among multiple robotic capsule modules. For example, active locomotion can be achieved via a collaborative actuation among multiple modules after self-assembly. This novel design was also supplemented with inductive power transmission techniques to wirelessly power the modules. The presented modular, miniature robotic platform may provide a new paradigm for developing multi-function capsule endoscope, with future implications in minimally invasive surgery. Keywords: Capsule robot, Modular robot, Endoscopy, Wireless power transmission. 1. INTRODUCTION Endoscopies are a standard medical procedure for diagnosing and treating abnormalities inside the GI track. They are more accurate than indirect methods, such as angiographies, ultrasonographies, and X-rays, in detecting abnormalities, because the doctor can directly view these areas using a wired endoscope with a small camera. An upper endoscopy is used to view inside of the upper digestive tract, including the esophagus, stomach, and first part of the small intestine, called the duodenum, and it enters the patient through the mouth. Meanwhile, a lower endoscopy, which is called a colonoscopy, is used to look at the large intestine (rectum and colon) and it enters through the patient's anus. During a colonoscopy, tissue samples can be collected (biopsy) and abnormal growths can be removed. However, the conventional endoscopy often requires sedation, because it may cause patients significant discomfort and intense pain, which may also increase the risk of intestinal perforation during the procedure. Furthermore, an endoscopy cannot inspect the small intestine, which is the middle portion of the gastrointestinal (GI) tract. Capsule endoscopy allows the doctor to view the small intestine in order to wirelessly diagnose bleeding and abnormalities, and it is painless unlike a conventional endoscopy [I]. There are commercial capsule endoscopy products for clinical use, including PiIICam (Yoqneam, Israel), Endo capsule (Olympus, Japan), MicroCam (IntroMedic, Korea), and SmartPill (SmartPill, USA) [2]. A typical endoscopy capsule is the size of a large vitamin tablet (approximately 11 mm in diameter and 27 mm in length) and contains a camera, LED lights, battery, and a telemetry module. The patient swallows the capsule with a glass of water and the camera inside the capsule captures and wirelessly transmits images to external data storage during its journey through the GI tract. The procedure is patient-friendly, because it is noninvasive and painless, and does not require sedation. Capsule endoscopy usually requires long examination times (approximately 6 8 hours) and after the procedure, doctors evaluate the suspected small intestine diseases based on the captured images [3]. Unfortunately, capsule endoscopy has limitations; for example, doctors cannot control the capsule because it moves passively through the natural bowel movement (peristalsis), which is caused by muscle contractions that ordinarily move food during digestion. As a result, the capsule cannot change its speed, position, or direction, and it cannot stay in a specific location for a more detailed examination. This passive movement can fail to detect suspicious lesions or to collect sufficient information about detected lesions. In addition, capsule endoscopy is only used for diagnosis and cannot perform biopsies or therapeutic procedures. If the abnormalities are identified, the capsule endoscopy should be followed by a conventional endoscopy m order to treat the abnormalities [4]. The integration of robot technology into capsule endoscopies could address some of these limitations. Capsule robots or robotic capsules have an active actuation mechanism so that they can move forward or backward along the GI tract and remain in position for a longer period of time. In addition, some capsule robots have been designed to perform medical functions, such as biopsies, drug deliveries, and surgical operations [5-7]. However, the single capsule robot has an intrinsic limitation: the small size of the single capsule robot limits the number of components and degree of freedom. Thus, complicated surgical operations cannot be performed using a single capsule robot. 350

2 Recently, a conceptual prototype of modular multiple capsule robots was introduced to address this limitation [8]. In the proposed system, tiny robotic modules are ingested separately and they assemble themselves into a specific structure in the stomach cavity in order to perform the surgical interventions. Based on the multiple modular robot approach, the paper suggests a novel modular robot system for diagnoses and interventions in the small intestine. We have also developed a prototype for the proof of concept. Our approach has several advantages compared to the single capsule robot for the small intestine. The proposed design is implemented through self-assembly, inchworm-like active locomotion via collaboration between the two capsule robots, a large stroke by the two capsule modules, and wireless power transmission for the power supply. The remainder of this paper is organized as follows: section 2 discusses the challenges of robotic capsule endoscopy, section 3 describes multiple capsule robots and section 4 presents our prototype, which consists of two robotic modules and a wireless power transmission. We conclude the paper in Section CHALLENGES IN ROBOTIC CAPSULE ENDOSCOPY Robotic capsule endoscopy requires technical improvement and needs to overcome several challenges before being used with patients. First, the size constraint of the capsule robot is a significant obstacle, because it highly increases the design difficulties. A large capsule causes patients discomfort and increases the possibility of retention inside the body. Commercial capsule endoscopies are sufficiently small for people to swallow; however, they are designed for diagnosis only and do not have therapeutic functions, such as microsurgery and drug delivery. Therefore, an effective capsule robot must include these functions in a single capsule while maintaining a sufficiently small size. Thus, an elaborated design and micromachining technologies should be considered. The second challenge is the locomotion mechanism that controls the capsule inside the body. Adjusting the speed and location of a capsule robot in the GI tract is very challenging due to the slippery and viscoelastic intestinal wall [9]. Based on several capsule robot prototypes, the locomotion mechanism can be divided into two approaches: internal and external actuation methods. In internal approaches, the capsule robot has a self-propelling mechanism that allows it to move actively in the GI tract. Dario et al. proposed legged locomotive mechanisms that mimic inchworms for colonoscopies, and these capsule robots are actuated using shape memory alloy (SMA) actuators and later using two micro-motors [10]. Kim et al. developed an earthworm-like capsule robot using DC motors and SMA actuators, and another paddling-based locomotive mechanism that originated from the motions used to paddle a canoe [11]. However, these active mechanisms have several limitations. The SMA actuator has very low efficiency and a slow response time due to its low heating and cooling speeds. Furthermore, the micro-motors require high power consumption and complicated design mechanisms. The second approach uses external actuation methods. External electromagnetic coils or permanent magnets generate strong magnetic fields to advance the capsule robot inside the body. Carpi et al. used an external magnet to move a capsule robot with a magnetic shell [12]. Sitti et al. suggested a magnetically deformed and actuated capsule endoscopic robot [13]. However, there are also obstacles with these actuation methods because strong magnetic fields are difficult to control and require a large and expensive facility. Another challenge for capsule robots is the power supply because the active locomotion requires a much higher power consumption to energize the actuators. The small volume of the capsule robot significantly limits the battery capacity; as a result, it limits the usage time and functionality of the capsule robot. In order to overcome this limitation, some researchers recently have successfully integrated a wireless power transmission (WPT) technology into capsule robots using inductive coupling [14, 15]. However, there are several issues with WPT as a platform for the capsule robot because the capsule robot is working inside the human body. First, WPT systems should provide a stable power supply regardless of the position and orientation of the capsule robot, which affects the efficiency of the power transmission. Second, the power should be transmitted over a long distance, usually more than 50 cm, between the transmitter outside the body and receiver inside the body. Third, human safety is a concern due to the electromagnetic field generated by the transmitting coil and high level of input power [19]. Ryu et al. suggested the coupling of a Helmholtz transmitting coil in order to generate a uniform electromagnetic field and a 3D receiving coil to produce stable power transmission for all orientations of a capsule endoscopy [16]. Puers et al. demonstrated that, regardless of capsule orientation, 300 mw can be transferred without time limitations [17]. However, the system requires a very high input voltage to the transmitting coil (up to 3.5 kv) due to high operating frequency of 1 MHz, which is not safe for humans. For higher efficiency over long distances, wireless power transfers based on the strongly coupled magnetic resonance theory can be used [18]. 3. MODULAR ROBOTS FOR MINIMALLY INVASIVE SURGERY Modular robotic platforms are a promising approach for minimally invasive surgery. The functional load for a procedure can be distributed among multiple robots, which address the size constraints of a single capsule robot. The assembly and cooperation among multiple robots allows the surgeon to perform complex surgical tasks. Assembling reconfigurable endoluminal surgical 351

3 (ARES) [9] system introduced a concept of modular robotics for endoscopic surgery. In this modular scheme, after ingestion, the miniaturized robotic modules assemble inside the stomach to build a specific structure. The assembled robots perform the target surgical operations in the stomach and then disassemble into individual modules or they can reconfigure themselves into another structure to perform other tasks. Researchers in this project implemented prototypes of these modular robots and demonstrated the assembly based on permanent magnets, but it remains a proof of concept. Nagy et at. introduced miniature robots for natural orifice surgery [20]. Natural orifice translumenal endoscopic surgery (NOTES) is a novel surgical approach that eliminates the need for external incisions by accessing the abdominal cavity through natural orifices. Surgical precision to perform complex surgical tasks is required during NOTES. However, this is not easily achieved, because NOTES currently has some limitations, including limited stability, triangulation, dexterity, and lack of adequate visualization, which result from the flexible endoscopy platform. In order to address these limitations, Nagy et at. use multiple independent miniature robots that can be inserted simultaneously into the abdominal cavity to provide a robotic platform for NOTES surgery. The robots are designed to provide enhanced visualization, better surgical dexterity, and improved triangulation for NOTES. 4. PROTOTYPE FOR ROBOTIC ENDOSCOPY IN THE SMALL INTESTINES 4.1 Design of the proposed modular capsule robots Based on the advantages of modular robots, we have developed a proof of concept for modular robots for use in diagnosis and therapeutic operations in the small intestine. Each capsule robot has four legs to anchor itself to the intestinal wall, a connector to connect to and then push/pull the other robot to create locomotion inside the GI tract, an RF receiver to communicate with an external transmitter, two micro linear servos for the legs and connector, and a power receiving module to wirelessly and continuously supply power to the payloads (see Fig. I). The housing of the capsules was fabricated in polylactic acid (PLA) using a 3D printer (MakerBot Replicator2). Robotic modules in each capsule are wirelessly controlled via a 2.4 GHz RF receiver and a transmitter. The dimension of the current capsule prototype is as large as a D-cell battery, which is approximately four times the size of a commercial capsule pill (see Figure l(b)). It will be miniaturized for being swallowed in the next stage of the development. 4.2 Active locomotion mechanism For the active locomotion, we implemented an inchworm-like locomotion based on the collaborative actuation between two capsule robots. Magnet (a) Rubber coating Fig. 1 (a) CAD design of the capsule robot (b) assembled two capsule robots. After being swallowed, the two capsule robots connect to each other inside the GI track (step 1-2 in Fig. 2). The docking mechanism, to connect the two robotic capsules, is achieved using permanent magnets. The penn anent magnets are attached at each end of the connectors and attract each other when two capsules are close enough. Once the two capsules are connected, the assembled robots then make the locomotion using a combination of opening/folding motions of the legs and pushing/pulling motions of the connectors between the two robots. Fig. 2 shows our locomotion mechanism used to advance along the GI track. In step 3, the first capsule folds its legs to make sliding on the intestinal wall easier. Meanwhile, second one opens its legs simultaneously and anchors itself to the intestinal wall. In step 4, the assembled robots push each other through connectors, resulting in the first capsule sliding forward while the second one is anchored. Then, the first capsule is anchored by opening its legs and the second one becomes ready to slide by folding its legs (step 5). Finally, the two capsules pull each other, as a result, the second capsule moves forward in step 6. By repeating these steps, the capsule robots can keep moving forward. The speed of advance can be adjusted by controlling the length and speed of the connector's movement. Similarly, the above steps can be done in the opposite direction if the assembled robot wants to move 352

4 backwards. In order to stay at an area of interest, both capsule robots open their legs to anchor themselves to the intestinal wall. Each leg is coated in a rubber material to improve the friction against the intestinal wall. maximum performance and the size limitation of the receiving coil. The power transmitter and receiver operated in a resonant state in order to improve the efficiency of the power transmission using capacitors on each side. The system was operated at a high resonant frequency of 7 MHz because the power transmission was more efficient at higher frequencies [19]. There are several advantages of our WPT system. The maximum transmitted power was 700 mw, which is sufficiently high to activate the capsule robots. The system is safe for patients, because it operates at a low input voltage (32 V). In addition, it is designed to be MRI compatible by using non-ferromagnetic materials, such as air core instead of a ferromagnetic core (see Fig. 3(b)). Helmholtz transmitting coils Transmitter system (a) Fig. 2 Inchworm-like locomotion based on a collaborative actuation between two capsule robots. 4.3 Wireless power transmission system Capsule robots require much more power than that supplied by an on-board battery. In our prototype, the system wirelessly transmits power to the two capsule robots based on the inductive coupling mechanism. The wireless power transmission (WPT) system consists of a power transmitter and receiver side. The power transmission system is equipped with a pair of Helmholtz transmission coils to generate a uniform magnetic field outside the human body, a timing generator to generate an alternating magnetic field through the Helmholtz coil, and a power amplifier. The power receiver system has a rectifier to convert the AC to DC and a regulator for continuous power supply to the capsules (see Fig. 3). The efficiency of the power transmission is dependent on the number of turns and cross sectional area of the receiving coil, the resonant frequency, and the density of the magnetic field. The transmitting (primary) copper coil is tightly wrapped with three turns around the acrylic ring, having a diameter of 43 cm. The size of the receiving (secondary) coil structure is 2 cm in diameter with a height of 1 cm. Receiving coils with 7, 14, and 42 turns were constructed and tested for the desired power transmission. The coil with 14 turns was selected for our prototype system considering both the Fig. 3 (a) WPT system overview, (b) transmitter system with Helmholtz transmitting coils (left) and receiver coil with 14, 28, and 42 turns (right). 5. CONCLUSION In this paper, we discussed the current state of capsule-based endoscopy research. Among the existing approaches, the micro modular robotic platform is a promising approach for minimally invasive surgery. Each capsule robot employs a simple locomotion mechanism by distributing the function loads among multiple robots. After the modular robots assemble inside the stomach to create a specific structure, they perform complex operations and locomotion. Considering these benefits, we designed and implemented a prototype of a novel modular capsule robot system for diagnoses and therapeutic operations in the small intestine. Two capsule robots work together to create an inchworm-like active locomotion by pushing and pulling each other. During the endoscopy, each capsule is powered wirelessly using inductive coupling between Helmholtz transmitting coils outside the human body and receiving coils inside the capsule robots. The wireless power system is safe for humans and is also MRI compatible. 353

5 This prototype modular capsule robot is in the early stage of development. The future research will focus on miniaturizing the capsule robot based on PCB fabrication and micromachining. In addition, we will verify and enhance the locomotion mechanism for the lubricated environment inside the GI tract, and improve the efficiency and stability of the wireless power transmission by finding the optimal transmitting parameters. REFERENCES [1] G. lddan, G. Meron, A. Glukhovsky, and P. Swain, "Wireless capsule endoscopy," Nature, Vol. 405, No. 6785, pp , [2] G. D. Meron, "The development of the swallowable video capsule (M2A)," Gastrointestinal Endoscopy, Vol. 52, No. 6, pp ,2000. [3] P. Swain, G. Tddan, G. Meron, and A. Glukhovsky, "Wireless capsule endoscopy of the small-bowel: development, testing and first human trials," Biomonitoring and Endoscopy Technologies, Vol. 4158, pp , [4] G. Pan, and L. Wang, "Swallowable Wireless Capsule Endoscopy: Progress and Technical Challenges," Gastroenterology Research and Practice, Vol. 2012, pp. 1-9,2011. [5] K. Kong, J. Cha, D. Jeon, and D. Cho, "A rotational micro biopsy device for the capsule endoscope," Proc. of the IEEEIRSJ International Conference on Intelligent Robots and Systems, pp ,2005. [6] S. Park, K. T. Koo, S. M. Park, and S. Y. Song, "A novel micro actuator for micro biopsy in capsular endoscopes," Journal of Micromechanics and Microengineering, Vol. 18, No. 2, pp , [7] 1. Wilding, P. Hirst, and A. Connor. "Development of a new engineering-based capsule for human drug absorption studies," Pharmaceutical Science and Technology Today, Vol. 3, pp , [8] The Assembling Reconfigurable Endoluminal Surgical System (ARES) Project website, Online Available: /fp6/nest /pdf/whats _next/ares. pdf. [9] J. S. Kim, 1. H. Sung, Y. T. Kim, E. Y. Kwon, D. E. Kim and Y. H. Jang, "Experimental Investigation of Frictional and Viscoelastic Properties of Intestine for Microendoscope application", Tribology Letters, Vol. 22, No. 2, pp ,2006. [10] M. Quirini, A. Menciassi, S. Scapellato, C. Stefanini, and P. Dario, "Design and Fabrication of a Motor Legged Capsule for the Active Exploration of the Gastrointestinal Tract ", IEEEIASME Transactions on Mechatronics, Vol. 13, No. 2, pp ,2008. [11] H. Park, S. Park, E. Yoon, B. Kim, J. Park, and S. Park, "Paddling based microrobot for capsule endoscopes," Proc. of IEEE Internatinal Conference on Robotics and Automation, pp ,2007. [12] F. Carpi, S. Galbiati, and A. Carpi, "Controlled navigation of endoscopic capsule: concept and preliminary experimental investigations ", IEEE Transactions on Biomedical Engineering, Vol. 54, No. 11, pp , [13] S. Yim and M. Sitti, "Design and Analysis of a Magnetically Actuated and Compliant Capsule Endoscopic Robot ", Prof of IEEE International Conference on Robotics and Automation, pp , [14] H. W. Li, G. Z. Yan, and G. Y. Ma, "An active endoscopic robot based on wireless power transmission and electromagnetic localization," International Journal of Medical Robotics and Computer Assisted Surgery, Vol. 4, pp , [15] B. Lenaerts and R. Puers, "An inductive power link for a wireless endoscope," Biosensors and Bioelectronics, Vol. 22, pp ,2007. [16] M. Ryu, 1. D. Kim, H. U. Chin, 1. Kim, and S. Y. Song, "Three-dimensional power receiver for in vivo robotic capsules," Medical and Biological Engineering and Computing, Vol. 45, No. 10, pp ,2007. [17] R. Puers, R. Carta, J. Thone, "Wireless power and data transmission strategies for next-generation capsule endoscopes," Journal of Micromechanics and Microengineering, Vol. 21, No. 5, pp , [18] X. Fang, H. Liu, G. Li, Q. Shao, and H. Li, "Wireless Power Transfer System for Capsule Endoscopy Based on Strongly Coupled Magnetic Resonance Theory ", Prof of IEEE International Conference on Mechatronics and Automation, pp , 20 II. [19] W. Xin, G. Yan, and W. Wang, "Study of a wireless power transmission system for an active capsule endoscope," International Journal of Medical Robotics and Computer Assisted Surgery, vol. 6, No. 1, pp , [20] Z. Nagy M. Fluckiger, R. Oung, I. K. Kaliakatsos, E. W. Hawkes, and B. J. Nelson, Assembling reconfigurable endoluminal surgical systems: opportunities and challenges, International Journal of Biomechatronics and Biomedical Robotics, Vol. 1, No. 1,