A magnetic internal mechanism for precise orientation of the camera in wireless endoluminal applications Authors P. Valdastri1, C. Quaglia1, E. Buselli1, A. Arezzo2, N. Di Lorenzo3, M. Morino2, A. Menciassi1, P. Dario1 Institutions 1 Center for Research in Microengineering (CRIM Lab), Scuola Superiore Sant’Anna, Pontedera, Italy 2 Centre for Minimally Invasive Surgery, University of Torino, Italy 3 Department of Surgery, University of Tor Vergata, Rome, Italy submitted 5 November 2009 accepted after revision 3 March 2010 Bibliography DOI http://dx.doi.org/ 10.1055/s-0029-1244170 Endoscopy 2010; 42: 481–486 © Georg Thieme Verlag KG Stuttgart · New York ISSN 0013-726X Corresponding author P. Valdastri, PhD CRIM Laboratory Scuola Superiore Sant’Anna viale R. Piaggio 34 56025 Pontedera Italy Fax: +39-0508-83497 pietro@sssup.it Original article 481 Introduction ! Although magnetic fields and their properties have been well known for centuries, their use in currently evolving surgical procedures is still lim- ited, even if rapidly increasing. A promising field of application is the external magnetic control of diagnostic or surgical devices deployed inside the patient’s body. In endoscopic procedures, this would allow maximization of the number of tools available with no effect on the number of access ports. In particular, magnetic camera systems can be introduced through natural orifices and oriented by magnetic forces. Several solutions describing this approach have already been reported [1–5]. However, the static magnet on the endoscopic or surgical device al- ways has a fixed orientation. Thus, the position and orientation of the device can only be con- trolled by varying the external magnetic field. In practice, this is obtained either by moving the ex- ternal magnets, if permanent magnets have been used for the external platform, or by adjusting the magnetic field strength if coils have been used to generate the magnetic field. This results in poor and difficult control of the remote device because of the exponential decrease of magnetic field strength with distance. Control is further affected by the unpredictable presence of different tissues in between the external and the internal magnet- ic sources, depending on the position of the de- vice inside the human body. Finally, because of the reliance on purely magnetic control, the re- mote device is maneuveredwith respect to an en- tirely external reference frame. In endoluminal cavities, this usually means that the surrounding collapsed tissues move with the device when the external magnetic forces are applied; adjustabil- Background and study aims: The use of magnetic fields to control operative devices has been re- cently described in endoluminal and transluminal surgical applications. The exponential decrease of magnetic field strength with distance has major implications for precision of the remote control. We aimed to assess the feasibility and functional- ity of a novel wireless miniaturized mechanism, based on magnetic forces, for precise orientation of the camera. Materials and methods: A remotely controllable endoscopic capsule was developed as proof of concept. Two intracapsular moveable permanent magnets allow fine positioning, and an externally applied magnetic field permits gross movement and stabilization. Performance was assessed in ex vivo and in vivo bench tests, using porcine upper and lower gastrointestinal tracts. Results: Fine control of capsule navigation and rotation was achieved in all tests with an external magnet held steadily about 15 cm from the cap- sule. The camera could be rotated in steps of 1.8°. This was confirmed by ex vivo tests; the mecha- nism could adjust the capsule view at 40 different locations in a gastrointestinal tract phantom model. Full 360° viewing was possible in the gas- tric cavity, while the maximal steering in the co- lonwas 45° in total. In vivo, a similar performance was verified, where the mechanism was success- fully operated every 5 cm for 40 cm in the colon, visually sweeping from side to side of the lumen; 360° views were obtained in the gastric fundus and body, while antrally the luminal walls pre- vented full rotation. Conclusions: We report the feasibility and effec- tiveness of the combined use of external static magnetic fields and internal actuation to move small permanent intracapsular magnets to achieve wirelessly controllable and precise cam- era steering. The concept is applicable to capsule endoscopy as to other instrumentation for laparo- scopic, endoluminal, or transluminal procedures. Valdastri P et al. A magnetic internal mechanism for precise orientation… Endoscopy 2010; 42: 481–486 D ow nl oa de d by : A lb er to A re zz o. C op yr ig ht ed m at er ia l. ity of the intracapsular magnet would allowmovement of the de- vice relative to the surrounding tissues, thus enabling effective and reliable controlled motion. In order to verify this principle, we developed a tethered endo- scopic capsule equipped with one internal electromagnetic motor that could rotate two permanent magnets, a real-time wired vision module, a battery, and an electronic circuit for the wireless control of the system. This device, the magnetic internal mechanism (MIM) capsule, works with an external static mag- netic field for both gross positioning and stabilization and fine control. In the present study, the MIM capsule concept and re- sults from ex vivo and in vivo validation trials are described for the first time. Materials and methods ! The MIM capsule body (●" Fig. 1a) is cylindrical, with a diameter of 15mm and a length of 48mm, and it incorporates one electro- magnetic direct current brushless motor (Namiki Precision Jewel Co., Tokyo, Japan), two small permanent magnets diametrically magnetized (N52 NdFeB, Supermagnete, Uster, Switzerland), a rechargeable battery (3.7 V, 20mAh, Plantraco, Saskatoon, Cana- da) and an electronic circuit for wireless motor control. A wired complementary metal-oxide-semiconductor (CMOS) camera (MO-S588-3T-N, Misumi, Taipei, Taiwan) with four white light- emitting diodes (LEDs) was used for the study because at present there is no wireless real-time compact camera on the market. Only the vision submodule is powered through thewired connec- tion; the intracapsular battery provides energy for the wireless motor and controllingmechanism. The external lens has a hydro- phobic coating to prevent moisture accumulation. The overall capsule weight is 14.4 g. The external magnetic field is provided by a cylindrical perma- nent magnet (N35 NdFeB, B&W Technology & Trade, China) (●" Fig. 1b). It is 60mm in diameter, 70mm in length and 1.5 kg in weight, and it is magnetized diametrically. The external mag- net is fixed in a passive hydraulic arm, which is manually con- trolled by themedical operator, similarly to an ultrasound device. A braking system allows the arm to be fixed in a given position by the user. The magnetic attraction force between the external and the internal magnets at a distance of 15 cm is 0.5 N, roughly cor- responding to 50 g, which is enough to lift the MIM capsule against gravity. Control software, running on a personal computer (PC), allows remote control of the device and feedback on its status, for exam- ple regarding battery level, wireless signal strength, etc. The user can rotate the camera view stepwise by pressing buttons on the PC keyboard, one for forward and one for backward motion. Alternatively, the rotation angle and direction can be set using the control software and an appropriate command is issued to the camera device. The internal orienting mechanism consists of a motor connected to a worm gear. When the worm screw rotates, it transmits the motion to a toothed gear. The small intracapsular magnets are fixed to the toothed gear and rotate when the motor is activated. real-time vision module electromagnetic motor battery and electronic circuit a permanent magnets Fig. 1 a Components of the magnetic inter- nal mechanism (MIM) capsule. b The MIM capsule and the per- manent magnet that provides the external magnetic field source. magnetic internal mechanism External magnet B Magnetic field N 15 cm N S S B Fig. 2 The working principle of the magnetic internal mechanism (MIM) capsule. When the motor is activated, the capsule rotates as the intracap- sular magnets stay in the same position with respect to the external mag- net. Original article482 Valdastri P et al. A magnetic internal mechanism for precise orientation… Endoscopy 2010; 42: 481–486 D ow nl oa de d by : A lb er to A re zz o. C op yr ig ht ed m at er ia l. However, when the system is placed in an external magnetic field, the magnets tend to maintain a precise alignment with re- spect to the external magnet, based on the position of their own north and south poles. If the magnetic force is enough to with- stand the torque of the motor, then the motor and the entire cap- sule rotates rather than the internal magnets which keep their orientation with regard to the external magnet (●" Fig. 2). This enables a fine adjustment of the camera relative to the surround- ing tissue, without moving the external magnetic source. ●" Video 1 shows an example of operation of the MIM capsule during bench testing, where the internal mechanism rotates the capsule back and forth by 90°. Lower and upper gastrointestinal phantom models The main goal of ex vivo tests was to assess the reliability and precision of movement provided by the MIM once the capsule had been navigated magnetically to the target location. Quantita- tive outcomes were the minimum angular increments through which the camera could be rotated, the maximum steering angle that could be obtained in the colon and in the gastric cavity, and the possibility of achieving a 360° sweep in the stomach. An upper gastrointestinal model with fresh porcine esophagus and stomach inserts, and a lower gastrointestinal model (●" Fig. 3a) with fresh porcine colon inserts attached along the fixtures, were used to simulate ex vivo the anatomical geometry of the digestive tract. Four pairs of red and black targets were stitched in random positions along the upper digestive tract, and the same number of targets was used for the lower gastroin- testinal experiment. Each red target was placed on the opposite side of the lumen to the corresponding black one. The external permanent magnet was fixed to a hydraulic passive arm in order to facilitate handling by the operator (●" Fig. 3b). The capsule was inserted through either themouth or the anus of the phantom and it was guided by means of the external magnet to visualize the targets. A constant endoluminal slight insuffla- tion, adequate for distension of the digestive tract, was applied. After each pair of targets had been approached, the external mag- net was kept in a fixed position and the internal mechanism was activated through the control software in order to precisely steer the field of view from the black to the red target and back again. Two sessions, one for the colon and one for the stomach, were carried out, with five trials for each target, for a total of 40 MIM activations. In vivo animal model After completion of the phantom trials, the MIM was assessed in an in vivo experimental session, with the aim of verifying the vision enhancement capabilities of the adjustable mechanism. The outcomes were confirmation of the capability of the MIM to move the view from one side to the other in the colonic lumen and to achieve a 360° panoramic view at three different locations in the gastric cavity. The feasibility study was done using two 30-kg female domestic pigs. The experiments were carried out in an authorized labora- tory, with the assistance and collaboration of a specially trained medical team, in accordance with all ethical considerations and the regulations related to animal experiments. After intravenous sedation and bowel preparation using water enemas, the capsule device was inserted into the colon through the anus and advanced for 40 cm using external magnetic guid- ance. The advance was observed frontally by means of a flexible endoscope (Karl Storz, Tuttlingen, Germany) introduced through a colotomy at about 50 cm from the anus and directed towards the device. This also allowed moderate distension of the bowel. During the advance of the device, the intracapsular mechanism was activated every 5 cm in order to orient the vision module to- wards the lateral walls of the lumen and to acquire a detailed view of the mucosa. When the target location at 40 cm was reached, the capsule was moved magnetically backwards to the anus to be retrieved. Fig. 3 a The lower gastrointestinal tract phantom set-up. b The phantom set-up and the external magnet on the hydraulic arm. Video 1 Operation of the magnetic internal mechanism (MIM). When the intra- capsular motor is activated, the entire capsule rotates as the intracapsular magnets retain their alignment with the external magnetic field. online content including video sequences viewable at: www.thieme-connect.de/ejournals/abstract/endoscopy/ doi/10.1055/s-0029-1244170 Original article 483 Valdastri P et al. A magnetic internal mechanism for precise orientation… Endoscopy 2010; 42: 481–486 D ow nl oa de d by : A lb er to A re zz o. C op yr ig ht ed m at er ia l. Similarly the capsule was introduced into the stomach through a duodenotomy, while a flexible endoscope was inserted through the esophagus to achieve an optimal view and to provide suffi- cient insufflation. The capsule was then moved around the gas- tric cavity by external magnetic guidance and held in three differ- ent regions of the stomach, i. e., the fundus, the body, and the antrum. For each location, we tried to achieve a 360° panoramic view by means of the intracapsular mechanism. Results ! Lower and upper gastrointestinal phantom models When each target area had been reached by gross magnetic navi- gation, the external magnet was fixed at 15 cm away the capsule. In these conditions, the MIM capsule was able to scan, moving from the black to the red target (●" Fig. 4) and vice versa, in mini- mum angular rotation steps of 1.8°. A typical target located 20mm from the camera was visualized in 2-mm increments. Re- garding the maximum achievable sweep angle, a 360° panoramic field of view was obtained in the phantom stomach, whereas in the colon the camera could be steered through no more than 45° in total due to collision with the lateral walls. Visualization and illumination, even if not comparable with that provided by standard flexible endoscopes, were sufficient for the sake of our tests. The telemetric link between the PC and the capsule was reliable up to a distance of 3m. Battery lifetime, with continuous opera- tion of the motor, was 25 minutes. The front lens did not need cleaning thanks to the hydrophobic coating of the glass. In vivo animal model Under in vivo conditions all the MIM functions performed cor- rectly, including the telemetric motor commands. Continuous images were captured and the entire surface of both the stomach and colorectal tract along which the capsule was advanced were successfully visualized. Fine orientation was achieved with the external magnet fixed 15 cm away from the animal skin. In all eight MIM activations along the 40-cm path in the colon, we were able to orient the camera from one side of the lumen to the other and back again, confirming the visualization in 2-mm steps. The trial in the colon was completed in 30 minutes and the battery was able to power the MIM for the entire duration. As regards the stomach trials, a 360-degree panoramic view (●" Fig. 5a– c,●" Video 2) was achieved in the fundus and in the body, while in the antrum complete revolution was blocked by the gastric walls. An example of the image stream from the cap- sule is shown in●" Video 3. The procedure time required to visu- alize the whole stomach surface was approximately 10 minutes. At the end of the entire trial, the capsule remained fully function- al. Gross inspection of the animals after euthanasia revealed no mucosal damage. Fig. 5 Endoscopic view of three different positions of the mag- netic internal mecha- nism (MIM) device, rotating about a point during the in vivo ex- periment in the gastric cavity. Fig. 4 View from the camera during lower gastro- intestinal phantom trials. Video 3 Video stream from the magnetic internal mechanism (MIM) on-board camera during the in vivo experiment in the gastric cavity. online content including video sequences viewable at: www.thieme-connect.de/ejournals/abstract/endoscopy/ doi/10.1055/s-0029-1244170 Video 2 Endoscopic view of the magnetic internal mechanism (MIM) capsule rotating about a point during the in vivo experiment in the gastric cavity. online content including video sequences viewable at: www.thieme-connect.de/ejournals/abstract/endoscopy/ doi/10.1055/s-0029-1244170 Original article484 Valdastri P et al. A magnetic internal mechanism for precise orientation… Endoscopy 2010; 42: 481–486 D ow nl oa de d by : A lb er to A re zz o. C op yr ig ht ed m at er ia l. Discussion ! In this study we introduce a novel concept for steering the video capsule camera, based on the combined use of an external static magnetic field and on-board actuation to move intracapsular permanent magnets. Since the distance between the external and internal magnets is kept constant during the operation, it is possible to achieve precise and reliable orientation of an endo- luminal wireless device. Transabdominal magnetic anchoring and guidance systems (MAGS) for minimally invasive surgery were first described in a laparoscopic setting by Cadeddu’s group [1], and more recent application to natural orifice transluminal endoscopic surgery (NOTES) has been reported by the same group [2]. The use of magnetic fields has been tested under NOTES conditions by other groups to enable liver retraction during cholecystectomy as well as to stabilize a mesh for implantation during abdominal hernia repair [3]. Cuschieri’s team used a ferromagnetic glue for gastric mucosa retraction in endoscopic surgery [4]. From the technical standpoint, the most advanced system to exploit magnetic fixa- tion and positioning for laparoscopic and NOTES procedures is the peritoneum-mounted imaging robot as reported by Oleyni- kov’s group [5]. This wired device contains a camera, an illumina- tion system, an electromagnetic motor, and two permanent mag- nets located in fixed positions at the two ends of the device. A magnetic handle on the patient’s abdomen is used to attract the magnets embedded in the robot. The handle can be moved across the exterior of the abdomen in order to position and pan the imaging device. The embedded motor is used to remotely tilt the camera to arbitrary angles. A similar solution is reported by Allen’s group [6], consisting of a 11-mm monoscopic insertable panning/tilting endoscopic imaging device. A magnetically guid- ed intra-abdominal camera for single-incision laparoscopic sur- gery (SILS) was recently demonstrated in humans by Cadeddu et al. [7]. Active capsule endoscopy is another field of research that is pres- ently exploiting the combined use of external and intracapsular magnets. In this case magnetic fields are used to stop and control the capsule, and to steer it towards areas of interest for diagnostic [8,9] and surgical [10] purposes. The efforts in this area by the two main commercial players in capsule endoscopy worldwide, namely Given Imaging [11] and Olympus [12], further attest to the promise of this approach. The current trend in diagnostic and surgical procedures is to minimize the patient’s operative trauma by decreasing as far as possible the number of external incisions while increasing the technical capabilities of endoluminal and transluminal instru- mentation. Therefore, given the poor controllability of magnetic fields, the development of miniaturized mechanisms that allow precise positioning and steering of the single device would im- prove the outcomes of both standard and innovative procedures, such as laparoscopy, NOTES, SILS, or capsule endoscopy. We en- visaged that magnetic properties could be used not only for re- traction and for fixation of the intra-abdominal tools, but also for precise steering and rotation. The novel feature of our prototype is the adjustable internal mag- netic mechanism that allows fine control of steering and rotation. Compared with the steering achieved by moving the external magnetic field source, the mechanism presented here provides higher precision because the distance between the external and internal magnets is kept constant during the operation. As demonstrated by our findings, the best results with the MIM capsule are obtained in large cavities, such as the stomach or the abdomen, where movement of the device is not hampered by luminal walls. Nevertheless, even in a constrained environment, the MIM capsule’s steering capabilities might be beneficial in en- hancing diagnostic outcomes. The MIM capsule design assumes a 15-cm distance between the external and internal magnets. With morbidly obese patients, such a value can easily be exceeded, and in such cases, a stronger external magnet can be used, by applying the selection criteria previously reported by our group [13]. It is worth mentioning that to be effective magnetic locomotion requires slight insufflation of the lumen [14]. In the case of gas- tric inspection, transparent liquid solutions can be used to achieve luminal distension [15]. Another promising approach using a wireless insufflation system for capsule endoscopy has been proposed [16]. Concerning power consumption, since the MIM is activated on demand by the user whenever steering of the camera is desirable, the precise battery lifetime cannot be predicted, but the user can take advantage of the battery level monitoring provided by the control software. While feasibility for diagnostic evaluation of the upper and lower gastrointestinal tract was demonstrated, limitations to the devel- opment so far require comment. First, further reduction in size is desirable and this can be achieved by decreasing the size of the actuation unit. This would allow the diameter to be less than 12mm, thus enabling introduction through a standard trocar for abdominal operation. Further work on the visualization system are also needed to improve image quality. This is substandard in the current prototype; however the visualization system was not the main focus of this work. Integration of a wireless camera is also one of the next steps, in order to address capsule endoscopy application and to achieve a completely wireless device. This ad- ditional improvement will raise a significant issue regarding power consumption. A possible solution for supplying all the in- tracapsular subsystemswithin a reasonable volumemay bewire- less power induction [17]. On the other hand, use of theMIM cap- sule for laparoscopy, NOTES or SILS [13] would allow the pres- ence of a tether. This wired connection could be used to power the system and to control both the camera and the MIM subsys- tems, thus solving any problem related to power consumption and allowing a drastic reduction in size, since battery and wire- less electronics would not be required anymore. Another crucial issue is the remote control by magnetic fields for navigation and coarse positioning. This problem may be addres- sed by adopting a robotic solution [14,18,19]. Regarding the study protocol, the evaluation was done in ex vivo and in vivo nonsurviving porcine models. While we believe that all areas of the tracts explored were visualized, this was not definitively proven. In addition, the study was not designed to distinguish normal from abnormalmucosal findings. The obvious constraint was the specific anatomy of the porcine colon, that forced us to restrict the test to the distal area. Nevertheless we believe that when the downscaled version of our prototype might be available and a better external fine-control system de- veloped and tested, further in vivo tests to prove feasibility and utility of a similar tool might be of interest, in view of the con- tinuing development of evermore active and less invasive tech- nologies for diagnosis and cure of digestive tract disease. Original article 485 Valdastri P et al. A magnetic internal mechanism for precise orientation… Endoscopy 2010; 42: 481–486 D ow nl oa de d by : A lb er to A re zz o. C op yr ig ht ed m at er ia l. Acknowledgments ! The authors are grateful to Mr. N. Funaro for manufacturing the prototypes and to Ekawahyu Susilo, Gastone Ciuti, and Carmen Cavallotti for their valuable support during fabrication and test- ing of the device. This material is based in part upon work supported by the Euro- pean Commission in the framework of the VECTOR (EU/IST- 2006-033970) and ARAKNES (EU/IST-2008-224-565) European Projects. Competing interests: None References 1 Park S, Bergs RA, Eberhart R et al. Trocar-less instrumentation for lapa- roscopy magnetic positioning of intra-abdominal camera and retrac- tor. Ann Surg 2007; 245: 379–384 2 Scott DJ, Tang SJ, Fernandez R et al. Completely transvaginal NOTES cho- lecystectomy using magnetically anchored instruments. Surg Endosc 2007; 21: 2308–2316 3 Ryou M, Thompson CC. Magnetic retraction in natural orifice translu- minal endoscopic surgery (NOTES): addressing the problem of trac- tion and countertraction. Endoscopy 2009; 41: 143–148 4 Wang Z, Wang L, Tang B et al. Retraction by surface ferromagnetisation of target tissues: preliminary studies on feasibility of magnetic retrac- tion for endoscopic surgery. Surg Endosc 2008; 22: 1838–1844 5 Lehman AC, Berg KA, Dumpert J et al. Surgery with cooperative robots. Comput Aided Surg 2008; 13: 95–105 6 Hogle NJ, Hu T, Allen PK, Fowler DL. Comparison of monoscopic inserta- ble, remotely controlled imaging device with a standard laparoscope in a porcine model. Surg Innov 2008; 15: 271–276 7 Cadeddu JA, Fernandez R, Desai M et al. Novel magnetically guided in- tra-abdominal camera to facilitate laparoendoscopic single-site sur- gery: initial human experience. Surg Endosc 2009; 23: 1894–1899 8 Carpi F, Galbiati G, Carpi A. Controlled navigation of endoscopic cap- sules: Concept and preliminary experimental investigations. IEEE Trans Biomed Eng 2007; 54: 2028–2036 9 Wang X, MengMQH. A magnetic stereo actuation mechanism for active capsule endoscope. Conf Proc IEEE Eng Med Biol Soc 2007; 2007: 2811–2814 10 Valdastri P, Quaglia C, Susilo E et al. Wireless therapeutic endoscopic capsule. Endoscopy 2008; 40: 979–982 11 Volke F, Keller J, Schneider A et al. In-vivo remote manipulation of mod- ified capsule endoscopes using an external magnetic field. Gastroin- test Endosc 2008; 5: AB121–AB122 12 Gheorghe C, Iacob R, Bancila I. Olympus capsule endoscopy for small bowel examination. J Gastrointest Liver Dis 2007; 16: 309–313 13 Simi M, Ciuti G, Tognarelli S et al. Magnetic link design for a robotic laparoscopic camera. Journal of Applied Physics 2010; 107: 09B302 14 Ciuti G, Donlin R, Valdastri P et al. Robotic versus manual control in magnetic steering of an endoscopic capsule. Endoscopy 2010; 42: 148–152 15 Tortora G, Valdastri P, Susilo E et al. Propeller-based wireless device for active capsular endoscopy in the gastric district. Min Invas Ther Allied Technol 2009; 18: 280–290 16 Toennies JL, Webster RJ, III. A wireless insufflation system for capsular endoscopes. ASME Journal of Medical Devices 2009; 3: 27514 17 Carta R, Lenaerts B, Thoné J et al. Wireless power supply as enabling technology towards active locomotion in capsular endoscopy. Biosens Bioelectron 2009; 25: 845–851 18 Ciuti G, Valdastri P, Menciassi A, Dario P. Robotic magnetic steering and locomotion of capsule endoscope for diagnostic and surgical endolu- minal procedures. Robotica. 2010; 28: 199–207 19 Carpi F, Pappone C. Magnetic manoeuvring of endoscopic capsules by means of a robotic navigation system. IEEE Trans Biomed Eng 2009; 56: 1482–1490 Original article486 Valdastri P et al. A magnetic internal mechanism for precise orientation… Endoscopy 2010; 42: 481–486 D ow nl oa de d by : A lb er to A re zz o. C op yr ig ht ed m at er ia l.