Mobile In Vivo Biopsy
and Camera Robot
Mark E. RENTSCHLER a,1, Jason DUMPERT a, Stephen R. PLATT a,
Shane M. FARRITOR a, Dmitry OLEYNIKOV b
aUniversity of Nebraska, Department of Mechanical Engineering
bUniversity of Nebraska Medical Center, Department of Surgery
Abstract. A mobile in vivo biopsy robot has been developed to perform a biopsy
from within the abdominal cavity while being remotely controlled. This robot
provides a platform for effectively sampling tissue. The robot has been used in vivo
in a porcine model to biopsy portions of the liver and mucosa layer of the bowel.
After reaching the specified location, the grasper was actuated to biopsy the tissue
of interest. The biopsy specimens were gathered from the grasper after robot
retraction from the abdominal cavity. This paper outlines the steps towards the
successful design of an in vivo biopsy robot. The clamping forces required for
successful biopsy are presented and in vivo performance of this robot is addressed.
Keywords. Mobile, Robot, Laparoscopy, In vivo, Biopsy
Minimally invasive surgery provides the patient with reduced trauma, but limits the
surgeon’s ability to manipulate the in vivo environment directly. The da Vinci surgical
robot system has been used to improve surgical dexterity. However, the tools utilized
by the da Vinci robot remain constrained by the entry incision. Tool changes still
require the removal of the existing tool and the reinsertion of the new one, adding to the
overall surgical time and adversely affecting the efficiency of the operation [1-2].
An alternate approach to external surgical robotics is to place the entire robot
within the abdominal cavity. These in vivo robots are unconstrained by the entry point
and can currently provide the surgeon with vision assistance . In vivo fixed base
camera robots have been developed that can provide the surgeon with additional views
of the abdominal environment . These additional views proved helpful for
determining orientation and defining depth during a porcine cholecystectomy . To
enhance the surgeon’s capabilities further a mobile robot was developed . This
wheeled robot has provided a mobile platform for an adjustable-focus camera system.
This mobile camera robot was used successfully to remove a porcine gallbladder
without the aid of a laparoscope . This procedure demonstrates the usefulness of in
vivo robotics by showing the practicality of a two-port cholecystectomy. The camera
port is eliminated by placing the camera inside the patient on an in vivo robot.
An in vivo manipulator robot could help the surgeon directly manipulate tissue. A
family of robots could be placed in vivo to perform a task collectively. The camera
robot could provide the visual feedback to the surgeon, while a grasper robot retracts
the tissue and third robot performs the dissection. To realize the full potential of in vivo
robotics, robots with grasping capability will be necessary.
1. Robot Design Requirements
There were three sets of design objectives for the biopsy robot. First, the robot would
need to provide sufficient clamping forces to sever hepatic tissue. Next, the robot
would need to provide sufficient traction to not only traverse the abdominal cavity, but
also enough to pull the sample free if not completely severed. Finally, the robot would
need to provide effective visual feedback for abdominal exploration and specifically
during the biopsy procedures.
The camera system for this robot builds on our previous work using camera
systems in vivo . The clamping and drawbar forces necessary for successful biopsy
were determined experimentally.
1.1. Clamping Force Measured
A biopsy forceps device that is commonly used for tissue sampling during esophago-
gastroduodenoscopy (EGD) and colonoscopies was modified to measure cutting forces
during tissue biopsy. These forceps, shown schematically in Figure 1, consists of a
grasper on the distal end with a handle/lever system on the proximal end. A flexible
tube is affixed to one side of the handle and the other end is attached to the fulcrum
point of the biopsy grasper. A wire enclosed in plastic inside the tube is used to actuate
the grasper. This wire is affixed to the free end of the handle lever and at the other end
to the end of the grasper lever arm. Actuation of the handle lever causes the wire to
translate relative to the tube and actuate the biopsy graspers. The tip of the forceps is
equipped with a small spike that penetrates the tissue during sampling. This normal
biopsy device was modified to contain a load cell to measure clamping forces indirectly,
as shown in Figure 1. Using this design, the force in the cable was measured.
Figure 1. Biopsy tool schematic with load cell in series with the actuation wire
Measurements of cable force were made while sampling liver, omentum, small
bowel and the abdominal wall of an anesthetized pig. Representative results for a liver
biopsy are shown in Figure 2 (left). The initial negative offset is due to the slight
compression in the cable to push the grasper jaws open before biopsy. The average
maximum measured force to biopsy porcine liver for three samples was 12.0 +/- 0.4 N.
These results are consistent in magnitude with other published results  concerning
forces sufficient to cut porcine liver.
Figure 2. Measured cable force to biopsy in vivo porcine hepatic tissue (left) and measured extraction force
to biopsy ex vivo bovine liver (right)
1.2. Extraction Force Measured
Generally, the biopsy forceps do not completely sever the tissue. In this case, the
forceps are gently pulled to free the sample. This extraction force needed to be
determined so that a robot could be designed to provide sufficient drawbar force.
A laboratory test jig was built to measure the force needed to free a biopsy sample
of bovine liver. After clamping the sample with the biopsy forceps a load cell attached
to the handle of the device was gently pulled to free the sample while the tensile force
was recorded. Representative results shown in Figure 2 (right) indicate that
approximately 0.6 N of force is needed to extract bovine liver tissue with the use of the
2. Robot Design
Based on the required clamping force and extraction force a successful biopsy robot
was designed, as shown in Figure 3. It provides a mobile in vivo platform for visual
feedback and for effectively sampling tissue. The wheels are independently controlled
by permanent magnet direct current motors to allow for forward, reverse and turning
motion. A linkage is used to actuate the biopsy grasper and focusing mechanism for the
camera. The robot’s grasper is 2.4mm wide and can open to 120 degrees.
Figure 3. Mobile in vivo camera and biopsy robot
3.1. Clamping Force Produced
Development of an actuation mechanism to drive the biopsy grasper and the camera
was achieved through several design iterations and testing. One challenge was
transforming the axis of rotation of the motor, which was perpendicular to the tube of
the forceps, to a line of translation parallel with the grasper wires.
The grasper is actuated using a lead screw linkage mechanism as shown in Figure
4. As the motor turns a lead nut is driven up and down. A linkage connects this lead
nut to the slider which is actuated horizontally. The grasper wires are attached to this
slider. The lead screw linkage mechanism was designed to maximize cable tensile force
as the mechanical singularity is reached during the actuation range of motion. The
camera lens was attached to this slider too. This provides the camera an adjustable-
focus feature necessary in the in vivo environment.
Figure 4. CAD drawing of the robot and actuation mechanism
Force measurements were made in the lab to determine the maximum amount of
force that could be produced using the biopsy robot design. Representative results from
these tests are shown in Figure 5 (left). The average maximum force produced for three
samples was 9.6 +/- 0.1 N. This force is about 16% smaller than the 12 N measured
during in vivo testing. However, the 12 N merely represents the force that was applied.
It does not represent the minimum force required to biopsy the tissue. It is probable
that the surgeon performed the biopsy and continued to increase the force and merely
“squeezed” the sample. The surgeon applied what he/she knew to be a sufficient force
rather than a minimum force. Also, the required force could also be largely reduced by
simply taking a smaller biopsy sample. Reducing the contact area by 16% would
produce the same applied stress.
Figure 5. Robot biopsy cable force production measured (left) and robot drawbar force production (right)
3.2. Drawbar Force Produced
As stated earlier a complete severing of the tissue is rarely achieved and some tearing
of the sample is usually needed to extract the sample. To be successful the in vivo robot
needed to produce enough drawbar force to pull the sample free. The biopsy robot
shown in Figure 3 was tested in vivo and with excised bovine liver to measure drawbar
forces. The tail of the robot was attached to a stationary load cell. The robot speed was
slowly increased as the drawbar force was recorded as shown in Figure 5 (right). After
maximum drawbar force was achieved, around 11 seconds, the robot wheel motion was
stopped. Results demonstrate that the robot is capable of producing approximately 0.9
N of drawbar force. This amount of force is 50% greater than the target of 0.6 N in the
laboratory measurements (Figure 2, right). This drawbar force is therefore sufficient
for sample extraction.
3.3. In vivo Results
In vivo mobility testing with this and other similar prototype robots, suggests that such
a wheel design produces sufficient drawbar forces to maneuver within the abdominal
environment. Recent in vivo porcine tests shows that the helical wheel design allows
the robot to traverse all of the abdominal organs (liver, spleen, small and large bowel),
as well as climb organs two to three times its height. These tests were performed
without causing any visible tissue damage.
The biopsy robot has been successfully tested in vivo in a porcine model. The robot
was first used to explore the abdominal environment while providing visual feedback to
the surgical team.
After exploring the abdominal environment, the biopsy mechanism was used to
acquire three samples of hepatic tissue from the liver of the animal. The robot camera
was used to find a suitable sample site. The biopsy graspers were opened and the
sample site was penetrated with the biopsy forceps’ spike. Then the graspers were
actuated. This cut nearly all of tissue sample free. The robot was then driven slowly
away from the sample site thereby pulling free the tissue sample. This tissue sample
was then retrieved after robot extraction through the entry incision. This demonstrates Download full-text
the success of a one-port biopsy and successful tissue manipulation by an in vivo robot.
Experiments were performed to determine the forces required to biopsy tissue. This
data lead to a successful mechanism design that is capable of producing sufficient
grasper forces to sample in vivo porcine tissues. A successful robot wheel design has
led to a robot capable of traversing the abdominal environment without causing tissue
damage. This wheel design is also capable of producing sufficient drawbar forces to
pull the biopsy sample free during in vivo testing. This robot design also incorporates
an adjustable-focus camera mechanism capable of providing visual feedback from
within the abdominal cavity of the patient.
Current work is focused on wireless developments and modification of the biopsy
forceps to provide a clamp capable of clamping shut a severed artery. These
developments are important for in vivo robotic use in forward environments such as
battlefield situations. These achievements in tissue manipulation and visual feedback
from within the abdominal cavity will further the development of in vivo robotics to
assist surgeons in forward battlefield situations and traditional medical centers.
 V. B. Kim, W. H. H. Chapman, R. J. Albrecht, B. M. Bailey, J.A. Young, L.W. Nifong, and W.R.
Chitwood, “Early Experience With Telemanipulative Robot-Assisted Laparoscopic Cholecystectomy
Using da Vinci,” Surgical Laparoscopy, Endoscopy & Percutaneous Techniques, vol. 12-1, pp. 33-44,
 H. Kang, and J.T. Wen, “Robotic Assistants Aid Surgeons During Minimally Invasive Procedures,”
IEEE Engineering in Medicine and Biology, vol. 20-1, pp. 94-104, 2001.
 Oleynikov, D., Rentschler, M., Hadzialic, A., Dumpert, J., Platt, S., Farritor, S., 2004, “Miniature
Robots Can Assist in Laparoscopic Cholecystectomy.” Journal of Surgical Endoscopy, 19: 473-476.
 Rentschler, M., Hadzialic, A., Dumpert, J., Platt, S., Farritor, S., Oleynikov, D., 2004, "In vivo Robots
for Laparoscopic Surgery." Studies in Health Technology and Informatics, 98: 316-322.
 Rentschler, M., Dumpert, J., Platt, S., Farritor, S., Oleynikov, D., 2005, “Toward In vivo Mobility.”
Studies in Health Technology and Informatics, 111: 397-403.
 Rentschler, M., Dumpert, J., Platt, S., Farritor, S., Oleynikov, D., “Mobile In Vivo Robots Provide Sole
Visual Feedback for Abdominal Exploration and Cholecystectomy.” J. Surgical Endoscopy – In Press.
 T. Chanthasopeephan, J.P. Desai, A.C.W. Lau, “Measuring Forces in Liver Cutting: New Equipment
and Experimental Results,” Annals of Biomedical Engineering, vol. 31, pp. 1372-1382, 2003.