Synopsis  

Colonoscopy clearly has a future, which will expand even if the technology stands still. There is a vast amount of colonoscopy to be done if recommendations for screening for colon cancer in everyone over the age of 50 were to be put into practice. However, there is something awful about colonoscopy. It hurts. Patients dread the procedure. Sometimes it is extremely difficult to do. Colonoscopists become inured to how terrible it sometimes is and simply say sorry as they push the colonoscope further into the patient who is already in pain. My three wishes for the future of colonoscopy are that:

• it should become painless
  
• sedation and analgesia should no longer be necessary
  
• it should become much quicker and easier.
  

There has been very little change in the nature of colonoscopy in the last twenty years. Despite some innovation, no substantial change altering the physical nature of colonoscopy has ever been tested in man. The advent of video colonoscopes, better application of stiffening including variable stiffness colonoscopes, magnifying images, and colonoscope magnetic localization systems have not altered the fundamental difficulty that is posed by colonoscopy. The most important limitation of colonoscopy is the tendency to form loops during advance through the colon. The loops are associated with increasing loss of transmission of force to the tip and consequent failure of advancement.

It is possible that colonoscopy might be overtaken by other technologies. These might include CT or MR colography (virtual colonoscopy) [1]. Screening for colon cancer might become unnecessary if fecal stool testing for genetic abnormalities or some form of screening blood test was found to be very specific. Some of these might reduce the anticipated volume of colonoscopy. Since MR and CT cannot see in color and cannot identify flat or very small abnormalities, these technologies seem unlikely to wipe out colonoscopy. However, these preliminary screening procedures might usefully increase the ratio of therapeutic to diagnostic colonoscopies if they can selectively identify which patients do not have polyps and therefore do not require colonoscopy. If an external imaging method such as MR colography could be found to image the colon reliably without preparation then the balance might tip substantially against diagnostic colonoscopy.

The main purpose of this chapter is to outline a variety of devices of varying practicality that have been described which might allow for easier examination of the large bowel. Some of these methods have been suggested for application at colonoscopy. These methods include tip propulsion by a variety of methods, robotics, wireless endoscopy, free capsule endoscopy, specialized overtube use, and toposcopy.

 Aids to advancing a colonoscope   

A variety of devices have been described that can be used in conjunction with an endoscope to facilitate its advance. These include overtubes, internal stiffening devices (spines), and guide threads or wires.

Overtubes   

A major problem for endoscopy of the small intestine (enteroscopy) is that, as the endoscope proceeds further and becomes more convoluted, so less and less of the force applied by the clinician is transmitted to the tip. It has been suggested that this problem might be eased by using an overtube, which can be slid over the endoscope to support it along its shaft so that the forces applied are restrained by the overtube without stretching the bowel into loops. Overtubes have been developed specifically to limit looping on the greater curve of the stomach.

Similarly, there has been some interest in designing special overtubes for colonoscopy. These have been tried in the past, and largely abandoned. The overtube can itself be an endoscope, as in the mother-and-baby system used for ileoscopy.

Effective overtubes are more difficult to design and construct than might be expected. The features that are helpful include a low coefficient of friction, a close fit between the endoscope and the tube, and increased stiffness in the sigmoid.

The ideal overtube would be floppy while it was being pushed into place over the endoscope and then become rigid so as to provide the best possible support. Bauerfiend et al. [2] have proposed an overtube with a hollow wall (Fig. 1) so that, when the air is sucked out of the annular gap, the inner and outer walls are squeezed together making it (moderately) rigid.

Internal spines   

An internal spine can be used instead of an overtube to alter the stiffness of the endoscope. One such device that has been used at enteroscopy is the stiffener made by Wilson Cook that utilizes the tendency of a wire-wound coil to become stiffer when it is compressed by tightening wires that run through the inside of the coil. This device can be passed through the biopsy channel of the colonoscope to increase its stiffness and may allow the tip of the endoscope to move further forwards when the scope is advanced at the anus. This internal stiffener becomes fixed within the channel of the endoscope, and they move as one.

Olympus have included a variable stiffness mechanism into a new colonoscope design. Some [3,4] but not all [5] clinical studies using this mechanism suggest that it may speed up colonoscopy.

The effectiveness of an internal spine might be enhanced if the colonoscope could be slid over the spine so that the spine served to guide it around curves. Sturges et al. [6] have proposed a 'slide motion' scheme where a flexible spine is slid forward a few centimetres out of the tip of the endoscope, it is then made rigid and the endoscope is advanced over the spine until the tip of the endoscope is adjacent to the tip of the spine. At this point the spine is once again made flexible and the cycle is repeated with the spine being slid forward again. To make a spine that can be switched rapidly between flexible and rigid they propose that the spine consist of a series of lose fitting balls and sockets which can be pulled into each other and locked when a wire that runs along the axis of the spine is tightened. They suggest that the tightness of the wire is controlled by passing an electric current through it so that its temperature and hence its length can be varied. This spine would be too large to fit through the biopsy channel of an existing endoscope and they envisage an 'endoscope conduit' which is a covering tube for the spine.

Mother-and-baby colonoscope systems   

A mother-and-baby ileoscopy system has been described by Jacobs [7]. A specialized mother colonoscope with a large channel was used to allow a baby flexible endoscope of 3.1 mm outer diameter with up and down deflection and a 1.2-mm channel to be used to intubate the ileum in 10 patients over a distance of 5–60 mm. This might be classified as an endoscope constrained within a stiff overtube.

Thread-guided pull endoscopy ('rope-way' colonoscopy and enteroscopy)   

Other endoscopic techniques have been used to evaluate the entire small intestine without surgery. A thread-guided method of enteroscopy is the oldest method to totally intubate the small intestine via the colon [8,9]. This technique involves having a patient swallow a guide thread and allowing it to pass through the whole gut until it emerges from the rectum. The thread is then exchanged for a somewhat stiffer Teflon tube over which an endoscope can be passed. In theory, a complete endoscopic examination can be obtained by a combination of pushing on the endoscope and pulling on the guide tube. The instruments are fully therapeutic including cauterization and polypectomy. Because the examination is painful due to tightening of the guide-tube, general anesthesia is usually required. This technique has been used with long flexible video endoscopes [10].

 Friction reduction   

Methods suggested for reducing friction include the use of silicone spray, vegetable oil spray, and even the use of a vibrating endoscope. Toposcopic self-everting catheters that unroll against the bowel wall have been suggested for following the lumen at endoscopy with a minimum of friction.

Lubricating the endoscope   

Friction can be reduced by using lubricants. Most endoscopists are familiar with the use of water-based lubricants such as KY jelly and use them to facilitate the introduction of gastroscopes or colonoscopes. Silicone spray was used for several years in endoscopy units but because of anxieties about possible toxicity (particularly if inhaled) there has been a trend away from using silicone towards using vegetable spray cooking oils or fats. Lubrication is particularly helpful when using enteroscopes in conjunction with overtubes. The overtube should be sprayed internally and externally from both ends and the endoscope should be sprayed as well. Routine lubrication of the shaft of the enteroscope is probably also helpful.

Vibrating the shaft of the endoscope   

Friction might be reduced by vibrating the endoscope. Hibino [11] has described an intricate endoscope, which contains components to make it vibrate. This device can be made to vibrate "in the vertically (upward/downward) or horizontally (rightward/leftward) directions, in the form of swing motion in which the distal end draws a circle, or in the form of movement (advance/retreat) motion."

Everting toposcopic endoscopy   

In 1978 Masuda [12] proposed that a flexible fiberscope could be fed through a conduit by attaching it to the end of an everted tube (i.e. a tube whose end has been turned inwards and pulled back through itself). When the tube is filled with liquid at pressure it will unroll itself and pull the endoscope forwards and, since the tube is rolling against the conduit wall, there is no sliding friction between it and the wall.

This technique has been used to pull catheters through vessels [13,14] and to carry an endoscope for falloposcopy [15].

 Novel propulsion systems   

The problems inherent to pushing a colonoscope have led to many proposals for methods of providing propulsion at the tip so that endoscopes can pull themselves rapidly through the gastrointestinal tract. Most of these methods are similar to the various ways that animals move and it is convenient to group them together under the headings of the relevant creatures.

Balloons to grip the wall—Earthworm   

Earthworms move by alternately extending and distending sections of their body to produce peristaltic waves that drive them through the soil. The most common approach to propelling endoscopes has been to imitate the motion of an earthworm by attaching inflatable segments to the endoscope. As embodied in Frazer's 1979 patent 'Apparatus for endoscopic examination'[16] (Fig. 2) there are two radially expandable bladders separated by an axially expandable bellows, with only the forward bladder attached to the endoscope. The sequence of operation is: (1) the rear bladder is expanded to anchor it against the colon wall; (2) the bellows are then expanded to push the front bladder (and hence the endoscope) forwards; (3) the front bladder is inflated so that it is locked in place against the colon wall then; (4) the rear bladder is deflated; and finally (5) the bellows are contracted to draw the rear bladder forwards ready to start the next cycle.

Variations on the worm theme can be found in several other patents [17–21]. The device of Liddy 1987 was the first worm method to be tested with humans. They used an overtube (42 in the patent drawing reproduced as (Fig. 3) to open or close the gap between the fore and aft balloons. It was tested in three patients with familial polyposis and apparently advanced well until, in every case, one or other latex balloon burst [22].

The methods described so far simulate an earthworm with only three sections, fore and aft sections that expand radially and a central section that moves axially. However, it is possible to make a more realistic worm with many sections. Such a worm is described by Grundfest et al. (Fig. 4) in a patent and an Internet article [21,23]. They use several segments so that waves of distension and extension can move along its body simulating a real worm. The distension is provided by rubber balloons which can be inflated to grip the bowel wall while small pistons provide extension. It has been tested in vivo in the small intestines of an adult pig which 'strongly resemble those of a human juvenile in size and mechanical properties'[23]. They report that results were encouraging and that substantial traction was possible but conclude: 'Although this machine could indeed move through a portion of the small intestine, it was clear that further development is required to support extensive in vivo experiments'.

Walter [24] described a double-balloon or inchworm colonoscope. To shorten and lengthen the colonoscope a push and pull flexible rod is used as a drive mechanism. A pneumatic cylinder is used to push the core in and out of the outer sheath.

Yamamoto has described a double-balloon method, mainly for enteroscopy, but he has used it in the colon [25]. This uses an 8-mm endoscope 200 cm in length with a balloon which can be inflated at its tip and an overtube with a balloon at its tip. The endoscope can be advanced and the balloon inflated to grip the intestine. The endoscope is then gently withdrawn, straightening the bowel. The overtube is advanced and the balloon on its tip is inflated. By repeating this cycle the bowel is pulled back and pleated over the overtube as the endoscope is advanced. The device requires two operators. It features pressure monitoring of the balloons in a control box. The whole procedure is performed under X-ray screening and an enteroscopy may take more than 2 h to perform. This is the first double-balloon system to have been used in patients. It represents a substantial advance in the technique of push enteroscopy since it can take therapeutic enteroscopes much further into the small intestine than hitherto. The biopsy channel size is only 2.2 mm which somewhat limits therapy but biopsy, cautery, polypectomy, and injection are possible. Experience in the colon with this device is extremely limited.

There has been interest in applying recent 'high tech' developments in micromachining and shape memory alloys to making more sophisticated worms. Guber [26] at the Karlsruhe Institut für Mikrostructurtechnik has proposed using minute valves and balloons to produce a worm to crawl through blood vessels while Carrozza et al. [27,28] have proposed a 'teleoperated' worm that has tiny robot arms to manipulate a video camera and take biopsy samples where required. They report that in vivo tests have been made and that 'the principle is suitable for propelling the microrobot in the colon efficiently without significant damage to the colon wall'.

Potential damage to the colon wall   

The issue of damage to the bowel wall is relevant since any earthworm system gets its grip by inflating a balloon that presses outwards against the walls and it has been known for many years that relatively small pressures can burst the colon and presumably the small bowel. In 1931 Burt [29] inflated the colons in a series of 18 cadavers and found that the pressure required to tear the serosa ranged from 43 KPa (325 mmHg) to only 5.4 KPa (41 mmHg) with a mean value of 18 KPa (137 mmHg). Balloon inflation has caused perforation of the small bowel during Sonde type enteroscopy [30].

Suction crawler—Limpet or starfish   

Suction can be used to grip the walls of the colon, which may have some advantages over using balloons, which tend to slip and may perforate the colon if overinflated. Carrozza et al. [31] have used a pair of suction heads separated by a bellows to move through lengths of excised porcine colon. The device looks very similar to the worm illustrated in Fig. 3, with the front and back bladders being replaced by suction heads which use an array of small suction holes to hold the wall, and the sequence of operation is the same as for the earthworm. The authors report that the 'prototype was able to navigate into the colon, both in the forward and backward directions, efficiently, consistently and at sufficient speed'. This group have more recently reported further development of a suction crawling device, which adds Velcro-like burrs to the suction heads [32]. Farhadi filed a patent on a somewhat similar idea using suction [33]. An internal tube is extended when a spring is unlocked and released. Suction is applied to the colon wall and a mechanism is applied which compresses the spring and locks it. By this mechanism, the endoscope tip is dragged forwards.

We have built a prototype system that is designed to fit onto an existing small diameter endoscope. In this system the tissue is gripped by the fore and aft suction heads, which are moved apart or together by a Bowden cable (a Bowden cable is a 'bicycle brake cable', i.e. an inner wire transmits force by sliding through an outer sleeve that is flexible but of fixed length). The sequence of operation is that the endoscope, with the front suction head extended, is inserted into the bowel and pushed in the conventional way, then, when progress becomes difficult, the suction in the front head is activated and once the tissue is gripped the head is retracted so that it pulls the tissues over the tip of the endoscope. The tissue is then gripped by the rear head and released by the front head, which is slid forward ready to repeat the cycle [34]. Experience with excised porcine colon arranged into tortuous curves showed that it was easy to advance through sigmoid bends that were difficult to traverse with a conventional instrument. We found that gripping the wall with suction caused no visible damage.

Vijayan [35] designed a hybrid balloon and suction device. This also used an extensor module sandwiched between two clamper modules. The clamper is a closed toroidal or doughnut-shaped balloon with six passive vacuum cups embedded onto a surface to give it a better grip. The air under the vacuum cups is squeezed out as the balloon expands against the bowel wall generating positive adhesion. The extensor module can then extend axially or change the direction of the robot colonoscope's tip

Serpentine robot—Snake   

Earthworms move by extension and distension, whereas many snakes rely on serpentine motion where 'the body literally swims along in a series of curves which gain a grip from exerting pressure against sticks, exposed roots, grass blades, pebbles, or slight irregularities in the ground'[36]. Robot snakes exist and at least one group has considered using them for endoscopy [6]. These authors rejected snake robots because they become 'computationally and mechanically burdensome as the number of degrees of freedom increases' and because it is difficult to miniaturize them sufficiently to be of use in endoscopy.

The mechanical aspects of this problem have been tackled by Ikuta et al. [37] who made an 'active endoscope' that is in effect a five segment snake. It uses shape memory alloy tendons arranged about a spine so that each section can bend in three dimensions. They show a series of pictures of it progressing along a rubber model of a section of bowel. In fact they do not operate it as an intelligent snake but rather use a joystick to manually control the two tip segments and the tip bending instructions are then passed back along the line as the endoscope is pushed forward so that subsequent sections follow their leader.

The computational aspects of making a snake have been addressed by Shan and Koren [38]. They made a simple snake that can move across a floor (i.e. in two dimensions) and is clever enough to move towards a planned position despite encountering obstacles.

Many legs—Millipede   

This is not quite an apt animal analogy but the principle is that an endoscope has many legs or rings around it that can be made to move back and forth and so march the endoscope forward.

Figure 5 is from Utsugi's patent [39] and shows the three inflatable cuffs that form one section of the millipede. The middle ('propellant') cuff is the leg which is pushed backwards and forwards by the cuffs either side of it. The sequence of operation is that the 'propellant' cuffs are inflated so that they press against the walls of the colon with enough force not to slip. Next, the 'drive' cuffs are inflated thereby pushing the propellant cuffs backwards so that the sheath, and hence the endoscope, moves forwards. The 'return' cuffs are now inflated so that they first lift the wall of the gut off the propellant cuffs and then push those cuffs back onto the drive cuffs which are simultaneously deflated. The cycle is now complete and one step has been taken.

Eleven years later Krauter described a somewhat similar but simpler method in his patent graphically titled 'Walking borescope'[40]. Four years later Krauter's colleagues at Welch Allyn [41] produced an ingenious design that used washers as feet.

In this design, illustrated in Fig. 6, the endoscope is surrounded by groups of five washers. All the groups are connected together and move in unison, but within each group every individual washer can be moved independently. Each of the five washers performs a cycle in which it moves slowly backwards and then rapidly forwards. If all five washers did this together then the endoscope would simply rock back and forth; but they don't because they are all out of phase so that at any one time four are moving slowly backwards and only one is moving rapidly forwards. As any driver can attest, the frictional force resisting skidding is independent of the speed of the skid, so in this case the forward propulsion from the four slow washers outweighs the reverse thrust from the one fast washer and the endoscope slowly advances.

Few legs—Lizard and ant   

Treat and Trimmer [42] present a four-legged device whose legs can extend as well as pivot at their proximal ends so that the quadruped can literally walk along the gut. It can be seen in Fig. 7 that the animal analogy is striking and that the creature has a single eye which sends a video image out through its tail to the endoscopist.

The article 'Future developments in high-technology abdominal surgery; ultrasound, stereo imaging, robotics'[43] shows a picture of a robot called Attila that looks like a giant ant and was designed for lunar exploration. The authors speculate that it might be possible to miniaturize this two kilogram robot and allow it to roam the gastrointestinal tract, but they do not review any of the problems that might arise and only conclude that 'the technology to do this is still not available'.

Water jet—Octopus   

The octopus escapes from predators by squeezing water from its mantle and jetting away. The physical principle is that the mass is accelerated by forcing fluid through a small orifice and the force required to do this produces a reaction that pushes the octopus in the opposite direction. It is the same principle that drives a rocket or a jet engine.

Our group [44] has developed a water-jet propelled endoscope. A spray head with a number of backward facing nozzles can be attached to an endoscope so that the endoscopist can use bursts of water to pull the endoscope out of a loop when further pushing only serves to enlarge the loop. We have found that it is practical to produce sufficient thrust without introducing an excessive amount of water in the bowel or an excessively sharp water jet. This propulsion force is helpful in advancing the endoscope along models made from plastic, rubber, and excised porcine colon as well as small bowel and has produced no visible damage in excised or in vivo pig colon.

Ginsburgh et al. in 1988 [45] proposed that this principle could be used to propel a 'borescope' for inspection of metal tubes such as drains or in other engineering applications.

Wheels and belts   

With the introduction of wheels and tracks the animal analogies must be dropped. Goh [43] has published a picture of a device that looks like a toy car and has shown photographs of a robot that is similar to a model of a First World War tank using miniature motors, which drive caterpillar tracks.

Takada [46] has patented a more conventional looking endoscope that has belts running along its shaft that are supposed to act like the tracks of a tank (Fig. 8). Little pulleys are used as drives for the belts. It is suggested that the grip from the belts onto the bowel wall will be sufficient to pull the colonoscope smoothly and painlessly into the patient.

 Wireless capsule colonoscopy   

Wireless capsule endoscopy has changed the way we think about enteroscopy. The system is strikingly successful in imaging patients with persistent gastrointestinal bleeding from the small intestine. Will wireless capsule technology [47–50] change the future of colonoscopy? There is little doubt that many patients fervently wish that a painless swallowed device might make the discomfort of colonoscopy a thing of the past. Good images are often obtained, especially of the right side of the colon. There are a number of technical issues, which require solution if wireless capsule technology is going to image the colon as well as it currently images the small intestine. Power management is a problem. With two small silver oxide batteries approximately 7 hours of imaging is obtained. The capsule usually takes much longer to pass through the whole colon, on average about 24 hours. More batteries, time delay and external transmission of power might solve the power problem. Methods of better preparation, feedback illumination, coping with intermittent rapid movements and prolonged periods of stasis, and better viewing of the whole of the mucosal surface of the colon, which is larger than the small intestine need to be found. Electostimulation has been used to move ovoid (capsule shaped) devices in the gut. Independent wireless devices have been used to drive experimental devices in the small intestine and colon [51].

 Future needs in other areas related to colonoscopy   

Sedation   

The response to sedation is variable. There is a need for better drugs than the opiate and benzodiazepine mixtures that are most commonly used. Propofol use is still controversial in colonoscopy. The rapid onset of deep sedation and quick recovery are desirable but its occasional severe suppression of respiration and hypotension are negative features. Its positive and negative features relate to its lack of protein binding unlike the other medications used. In the future we need better and safer sedatives.

Bowel preparation   

For some patients, bowel preparation can be very violent and uncomfortable. For the colonoscopist the preparation is frequently inadequate. Preparation in patients with severe bleeding is often difficult.

Instrument disinfection   

Cleaning colonoscopes adequately still remains a concern. One potential future mishap waiting to occur might be a small epidemic of hepatitis C, or HIV associated with poor cleaning of endoscopes. Publicity about such events might impact very seriously on the practice of colonoscopy. In the UK colonoscopes and all colonoscopies in patients are tracked to assess the possible transmission of prion ('mad cow') disease. Any suggestion that colonoscopy could transmit prion disease might wipe colonoscopy out or change its nature substantially. It is not hard to predict that the red tape and form filling which has become an increasing and dreary aspect of endoscopy will continue to increase in the future.

New imaging methods   

It is hard to predict the impact that optoelectronic aids to colonoscopy will have in the future. Magnification, optical coherence tomography, and various types of spectroscopy may alter the practice of diagnostic colonoscopy. Most of these relate to taking the diagnosis of cancer or neoplasia from the histology laboratory to the tip of the endoscope. It is probable that imaging methods which can visualize structures in or beyond the wall of the colon will become an increasing part of colonoscopy. Endoscopic ultrasound in the colon will be used more commonly to stage tumors and direct therapy.

 The future of therapeutic colonoscopy   

The future of flexible endoscopic surgery in the colon depends on the development of better surgical tools. The development of transanal miscrosurgical techniques which allow full-thickness excision of large villous adenomas and cancers with subsequent stitched closure of the defect has been shown to improve outcomes when compared with conventional surgical approaches. The development of safe and effective methods to perform such surgery at flexible colonoscopy is feasible. It is likely that submucosal resection of tumors will be more widely practiced and that as tools and techniques improve, full-thickness resection with closure of the defect and even anastomosis will become routine. Better sewing and stapling methods for use during flexible colonoscopy are needed.

 Conclusion   

The contents of this chapter might be a junkyard of ideas about improvements to colonoscopy, which will be rapidly discarded and forgotten. It could be that in the future colonoscopy will change very little. There are things about the procedure which stimulate the imagination to hope for better methods of colonoscopy in the future.

 Acknowledgments   

The author would like to thank Sandy Mosse, Tim Mills, Feng Gong, Gary Long, and Annette Fritscher-Ravens for their help in developing the ideas expressed in this article.

 References  

  1 Ahlquist, DA, Hara, AK & Johnson, CD. Computed tomographic colography and virtual colonoscopy. Gastrointest Endosc Clin N Am 1997; 7: 349–5.

  2 .Bauerfiend, P & Bauerfiend, H.Tubular Inserting Device with Variable Rigidity. US patent 5, 337,733 1994.

  3 Shah, SG, Brooker, JC, Williams, CB, Thapar, C & Saunders, BP. Effect of magnetic endoscope imaging on colonoscopy performance: a randomised controlled trial. Lancet 2000; 356 (9243): 1718–22. PubMed

  4 Brooker, JC, Saunders, BP, Shah, SG & Williams, CB. A new variable stiffness colonoscopes makes colonoscopy easier: a randomised controlled trial. Gut 2000; 46: 801–5. PubMed

  5 Rex, DK. Effect of variable stiffness colonoscopes on cecal intubation times for routine colonoscopy by an experienced examiner in sedated patients. Endoscopy 2001; 33: 60–4. PubMed

  6 Sturges, RH & Laowattana, S. 1991 A flexible tendon-controlled device for endoscopy. Proceedings of the 1991 IEEE International Conf on Rob and Automation, Sacramento, 2582–259.

  7 Jakobs, R, Benz, C, Maier, M, Kohler, B & Riemann, JF. Transvalvular enteroscopy using a mother-baby endoscope system: a new approach to the distal ileum. Gastrointest Endosc 1997; 45: 291–5. PubMed

  8 Deyhl, P, Jenny, S & Fumagalli, J. Endoscopy of the whole small intestine. Endoscopy 1972; 4: 155–7.

  9 Classen, M, Fruhmorgan, P & Kock, H. Peroral enteroscopy of the small and large intestine. Endoscopy 1972; 4: 157–62.

 10 Sato, N, Tamegai, Y, Yamakawa, T & Hiratsula, H. Clinical applications of the small intestinal videoendoscope. Endoscopy 1992; 24: 631 (Abstract).

 11 Hibino, H, Nagayami, Y & Yoshikawa, M, et al.Endoscope Apparatus. US patent 5, 060,632 1991.

 12 Masuda, S. Apparatus for Feeding a Flexible Tube Through a Conduit. UK patent 1, 534,441 1978.

 13 Shook, DR, Doppman, JL, Cattau, EL & Goldstein, SR. Everting (toposcopic) catheter for broad clinical application. Trans ASME, J Biomech Eng 1986; 108: 168–74.

 14 Benjamin, BS & Collins, KS. The toposcopic everting through-lumen catheter to facilitate dilation of severe strictures of the GI tract. GI Endosc 1986; 32: 33–5.

 15 Bauer, O, Diedrich, D & Bacich, S, et al. Transcervical access and intra-luminal imaging of the fallopian tube in the non-anaesthetized patient; preliminary results using a new technique for fallopian access. Hum Reprod. 1992 Suppl 1:7–11.

 16 Frazer, RE.Apparatus for Endoscopic Examination. US patent 4, 176,662 1979.

 17 Krasner, LK & DiBenedetto, JP.Medical Apparatus Having Inflatable Cuffs and a Middle Expandable Section. US patent 4, 676,228 1987.

 18 Shishido, Y, Adachi, H & Hibino, H, et al.Pipe Inspecting Apparatus Having a Self Propelled Unit. US patent 5, 090,259 1992.

 19 Krüger, PC.Insertion Assisting Device for Medical Instrument. US patent 5, 454,364 1995.

 20 Liddy, J, Sugarbaker, PH & Penland, WZ.Medical Apparatus. US patent 4, 690,131 1987.

 21 Grundfest, W, Slatkin, BA & Burdick, J.Robotic Endoscopy. US patent 5, 337,732 1995.

 22 Sugarbaker, PH, Penland, WZ & Liddy, J. Pneumatic device for the automated advancement of the fiberoptic endoscope for total colonoscopy. Gastrointest Endosc 1985; 31 (3): 210–3. PubMed

 23 .Slatkin, BA, Burdick, J & Grundfest, W.The Development of a Robotic Endoscope. Department Mech Eng. 1996, Caltech, brett@robby.calTechnical edu.

 24 Walter, WW, Sanjiv, H & Irwan, MK. Self-propelling colonoscope. [Online]: http//http://www.rit.edu/jnantl/asmeabs.html.

 25 Yamamoto, H, Sekine, Y & Sato, Y, et al. 2001 Total enteroscopy with a nonsurgical steerable double-balloon method. Gastrointest Endosc; 53 (2): 216–20. PubMed

 26 Guber, AE. Potential of microsystems in medicine. Minimally Invasive Ther 1995; 4: 267–75.

 27 Carroza, MC, Dario, P, Lencioni, B, Magnani, B, Reynaerts, D, Trivella, MG & Pietrabissa, A. (1996) .A Microrobot for colonoscopy. Proceedings of the of Seventh International Symp on Micro Machine and Human Science. IEEE, pp. 223–228

 28 Dario, P, Carroza, MC, Lencioni, B, Magnani, B & D'Attanasio, S. (1996) A miniature robotic system for teleoperated colonoscopy. Proceedings of the of IARP Workshop on Medical Robotics. Vienna October 1996.

 29 Burt, CAV. Pneumatic rupture of the intestinal canal with experimental data showing the mechanism of perforation and the pressure required. Arch Surg 1931; 22: 875–902.

 30 Morris, AJ & MacKenzie, JF. Small-bowel enteroscopy and NSAID ulceration. Lancet 1991; 337 (8756): 1550. PubMed

 31 Carroza, MC, Dario, P, Lencioni, B, Magnani, B, Reynaerts, D, Trivella, MG & Pietrabissa, A. The development of a microrobot system for colonoscopy. Editted by Troccaz J, Grimson E, Mosges R. Proceedings of the of 1st Joint Conf Computer Vision, Virtual Reality and Robotics in Medical and Medical Rob and Comp Assisted Surgery, pp. 779–88. Springer,.Grenoble.

 32 Menciassi, A, Arena, L & Phee, D, et al. (2001) Locomotion issues and mechanisms for microrobots in the gastrointestinal tract. 32nd International Symposium on Robotics, pp. 428–32.

 33 Farhadi, A.Creeping Colonoscope. US patent 5, 309346 2001.

 34 Mosse, CA Endoscopic propulsion mechanisms [Online]http://www.medphys.ucl.ac.uk/research/mle/endoscopy/propulsion.htm

 35 Vijayan, KA, Sanjiv, K & Irwan, MK. A fully automonomous microrobotic endoscopy sysytem. J Intelligent and Robotic systems Netherlands 2000, pp. 325–41.

 36 Carr, A. (1964). Life Nature Library. The Reptiles. Time Inc.

 37 Ikuta, K, Masahiro, T & Hirose, S. (1988), Shape memory alloy servo actuator system with electric resistance feedback and application for active endoscope. IEEE International conf on Robotics and Automation427–30.

 38 Shan, Y & Koren, Y. Design and motion planning of a mechanical snake. IEEE Trans on systems. Man, Cybernetics 1993; 23 (4): 1091–100.

 39 Utsugi, M.Tubular Medical Instrument Having a Flexible Sheath Driven by a Plurality of Cuffs. US patent 4, 148,307 1979.

 40 Krauter, AI.Walking Borescope. US patent 4, 934,786 1990.

 41 Allred, JB IIISelf Advancing Endoscope. US patent 5, 345,925 1994.

 42 Treat, MR & Trimmer, WS.Self Propelled Endoscope Using Pressure Driven Linear Actuators. US patent 5, 595,565 1997.

 43 Goh, P, Tekant, Y & Krishnan, SM. Future developments in high technology surgery: ultrasound, stereo imaging, robotics. Bailliere's Clin Gastroenerol 1993; 7: 961–87.

 44 Mosse, CA, Mills, TN & Swain, CP. A water jet propelled colonoscope (abstract). Gut 1997; 41 (Suppl. 3) E2.

 45 Ginsburgh, I, Carlson, JA, Taylor, GL & Saghatchi, H.Method and Apparatus for Fluid Propelled Borescopes. US patent 4, 735,501 1988.

 46 Takada, M.Self Propelled Colonoscope. US patent 5, 562,601 1996.

 47 Gong, F, Swain, CP & Mills, TN. An endorobot for gastrointestinal endoscopy. Gut 1994; 35: S52. PubMed

 48 Iddan, GJ & Sturlesi, D.In Vivo Video Camera. US patent 5, 604,531.

 49 Gong, F, Mills, TN & Swain, CP. Wireless Endoscopy Gastrointest Endosc 2000; 51: 725–9. PubMed

 50 Iddan, G, Meron, G, Glukhovsky, A & Swain, P. Wireless Capsule Endoscopy Nature 2000; 405: 417. PubMed

 51 Mosse, CA, Mills, TN, Appleyard, MN, Kadirkamanathan, SS & Swain, CP. Electrical Stimulation for propelling endoscopes. Gastrointest Endosc 2001; 54: 79–83. PubMed

Copyright © Blackwell Publishing, 2004