MECHATRONIC SYSTEM FOR TRANSURETHRAL

RESECTION TRAINING

Ángel Asensio, Alejandro Ibarz, Jose Ignacio Artigas, Álvaro Marco

Javier Casas and Roberto Casas

Instituto de Investigación en Ingeniería de Aragón, Universidad de Zaragoza, María de Luna 1, Zaragoza, España

Keywords: Transurethral resection, TUR, endoscopic operation.

Abstract: Training the residents who start with endoscopic operations remains a challenge. This paper describes an

electromechanical system developed for learning the transurethral resection (TUR) technique. This system

can be easily set and cleared up in a classroom, and consists of a supervisor’s workbench with a wireless

sensing device, connected to several trainees’ workbenches with motorised devices. These devices have a

resectoscope mounted on an electromechanical structure that is able to reproduce all the movements of an

actual endoscopic operation of the prostate.

1 INTRODUCTION

Transurethral resection (TUR) is an endoscopic

surgical technique that makes it possible to extract

tissue from the prostate in mitigating or corrective

operations. It is performed by means of a

resectoscope, consisting of a thin cannula that

includes an endoscopic lens system and contains a

tiny wire loop acting as an electro-scalpel, and

operated from the exterior.

Nowadays, video-surgery has simplified TUR

training, becoming a common practice among many

urologists. However, no urologist hesitates to

consider learning this technique difficult and time-

consuming. Mere vision of the moving endoscopic

image is not enough to acquire the reflexes, manual

skill, and mental agility necessary to cope with the

recurring occasions in which only solid practical

experience will make it possible to handle the

situation, and conclude the operation successfully

(Pycha, 2003).

Traditional training for novel surgeons is carried

out by first explaining the techniques with

endoscopic images, to later begin performing very

simple operations, directly in the operating theatre.

Work has been done on support for training in

endoscopic techniques, and there have also been

efforts to devise manipulators for motorised

operations guided by the surgeon (Kerfoot, 2004;

Gettman, 2003; Katz, 2003; Ottensmeyer, 2000;

Ballaro, 1999; Gomes, 1999). However, these

systems have certain problems, such as lack of

tactile feedback, and their high cost for generalised

use in training. Their main objective is to automate

operations, not to perform exact imitations of the

movements of a surgeon. Our system achieves a

great precise reproduction of the movements of an

expert surgeon, at a much lower cost.

Another difference with the mentioned line of

work is that our proposal aims at low-cost robot

systems, specific for this operation, capable of

capturing the movements of the resectoscope and

reproducing them both in real time and recorded.

For this, we have devised an easy to set and clear

up lecture room, practical for use at hospitals. The

room has a sensing workbench connected to a

computer, and several motor workbenches linked by

Bluetooth. All these workbenches are able of

reproducing the movements of a hand at the degrees

of freedom of the resectoscope. A video monitor

shows images of an operation. The sensing

workbench senses the movements performed upon

it, and the motor workbenches are able of

reproducing these movements.

The solution we present in this paper is cost-

effective, and has been successfully tested by

experienced surgeons. First we detail the specific

goals pointed out by the users, which guided us in

the design of the solution presented; Next, we

analyze in depth each block of the final system.

Asensio Á., Ibarz A., Ignacio Artigas J., Marco Á., Casas J. and Casas R. (2008).

MECHATRONIC SYSTEM FOR TRANSURETHRAL RESECTION TRAINING.

In Proceedings of the First International Conference on Biomedical Electronics and Devices, pages 74-81

DOI: 10.5220/0001051300740081

 SciTePress

2 SYSTEM GOALS

The aim of our system is to aid learning of a

complex technique of endoscopic surgery,

transurethral resection, in order to improve the skill

of surgeons and reduce risks for patients. We try to

offer a lecture room that permits training in

techniques as they are currently performed, in such a

way that its structure allows inclusion of more data,

and with feasible installation in hospitals (easily set

and cleared up in multipurpose halls).

Thanks to the system’s modular structure,

several different teaching modes are possible:

1. Real time teaching mode: The trainer operates

the sensing workbench and movements are repeated

in the hands of trainees by means of the motor

workbenches. The trainer’s explanations can be

underlined by videos played on the computer.

2. Recording mode: The trainer watches an

endoscopic video on the computer, and

simultaneously performs the corresponding

movements with the sensing workbench. The

movement pattern is stored in the computer, and a

video of the operation with embedded information

on positions is generated. This video will be used

later to control motor workbenches without the

presence of the trainer.

3. Recorded teaching mode (without the

presence of the trainer): The computer plays a pre-

recorded video that controls the motor workbenches.

4. Trainee assessment without trainer: A video is

played and trainee operates the sensing workbench;

movements are stored in the control board of the

sensing workbench. Once the test has ended, data

are sent to the trainer’s computer, and they are

checked against the movement pattern of the expert

surgeon.

The technical requirements we have had to deal

with are the following:

– The workbenches must reproduce movements

with the same degrees of freedom that the surgeon

has during an actual operation. The initial hypothesis

is that the urinary tract sphincter is fixed in space,

which leads us five degrees of freedom. The

correctness of this hypothesis and the validation of

the movement replicator have been checked during

the research actions performed by the group.

– Another very important requirement is

synchronization of the endoscopic video playing

with the movements of the resectoscope.

– To ensure portability, we envisage a

radiofrequency workbench data communications

structure using Bluetooth (Anastasi, 2003).

Bluetooth chips available on the market managed

from a microcontroller based system have been

used. Both the hardware and the firmware have been

original developments aimed at the present final

application.

– One last goal is to allow collection of actual

movements data in the operating theatre. In this

scenario it is not possible to modify the instruments

available to the surgeon, nor interfere with his

movements. To achieve this, we propose a new

method for sensing and capture of the movement of

the resectoscope in the operating theatre, based on

ultrasound.

3 DESCRIPTION OF THE

ENVIRONMENT

3.1 Basic Scenarios

Next we briefly describe the structure of a possible

lecture room. It should be pointed out that some

systems share certain common blocks, so these will

be described only once.

3.1.1 Trainer’s Workbench

Its role is that of a general coordinator, and it

comprises several clearly distinct subsystems

(Figure 1).

The resectoscope or instrument to be used by the

medical personnel is mounted on a mechanical

system that allows mobility as if an operation were

being performed. For this, three turns (coordinate

axes) and two sweeps (cannula and resection loop)

are allowed. The trainer will introduce the sequence

of movements using this workbench. A set of

position encoders capture kinetics directly, or by

means of the corresponding transmission ratios.

A digital system based on a Field Programmable

Gate Array (FPGA) has been developed for data

collection tasks, management of communications,

motor device control and memory management.

Previous developments were based on

microcontroller solutions, but the large number of

inputs-outputs and the need for concurrence

recommended migration to programmable logic

devices. The use of FPGA allows for modular and

flexible design, which eases the integration of the

various subsystems developed.

A Bluetooth device in this workbench acts as a

master of the wireless communications system. It is

possible to control a complete network, commanding

the various devices and modes, and at the same time

the various flows of information.

MECHATRONIC SYSTEM FOR TRANSURETHRAL RESECTION TRAINING

Acting as a central server, the trainer’s

workbench relies on computer equipment offering

various functions: It controls reproduction of the

TUR operation video; it allows storage of movement

patterns for later analysis or repetition; and it

manages communications and state of the devices in

the lecture room by means of Bluetooth linking via

electronic system.

Figure 1: Trainer’s workbench.

3.1.2 Trainee’s Workbench

The mechanical structure is similar to that of the

trainer’s workbench (Figure 2). It includes the

electromechanical devices allowing reproduction of

movements. The student holds the device and feels

the movement to be performed. The device retains

the position encoders, making available a process of

auto-calibration without the need of supervision by

the user. A Bluetooth device acts as a slave in the

network managed by the trainer’s workbench.

Figure 2: Trainee’s workbench.

3.1.3 Installation in Operating Theatre

Figure 3 shows the block diagram of the installation

in operating theatre. Encoder-based position capture

is not possible in the operating theatre, since the set

of instruments cannot be modified (figure 4). We

have therefore developed a novel positioning

technique based on ultrasound (US) pulses, with the

aim of applying it to the capture of the movements

performed by the surgeon on the resectoscope in the

operating theatre. In this way, it will be possible to

document fragments of actual operations with the

video information, movements, and other parameters

that may be considered relevant. We should point

out that the training video is obtained by an

endoscopic camera during the operation. Later on,

and as a previous step to its use in the training

system, it is processed by the computer system.

Figure 3: Installation in operating theatre.

3.2 Mechatronic System

Initially, the typical movement of the resectoscope

in an operation was studied. Considering the results,

to express the movement as parameters, the

following hypothesis is adopted: “The point of the

resectoscope oppressed by the sphincter is

considered still, and this point does not vary

throughout the operation”.

To all practical effects, it is considered as the

origin of the coordinates of the mechanical system.

On the basis of this hypothesis, the conclusion

reached is the need to design a mechanical system

allowing three angular movements, one forward

movement for the cannula, another for the cutting-

loop, and two for the switches of irrigation and

coagulation. To sum up, seven coordinates are used

to define the state of the device at a given moment.

Four of them reflect the point in space where the end

of the resectoscope is, another the state of the cutting

loop, and the remainder refer to the state of the

switches the resectoscope is equipped with

(generally pedal-operated).

BIODEVICES 2008 - International Conference on Biomedical Electronics and Devices

According to the explanation offered, the

mechanical system should allow the movements

detailed in Figure 4.

Turn 3

Turn 2

Turn 1

Extreme of

resectoscope

Turn 3

Turn 2

Turn 1

Extreme of

resectoscope

Figure 4: Kinetic diagram of the resectoscope.

The final system has been the result of several

different prototypes. At the moment of designing the

mechanical system, we have two different kinds of

devices:

Sensing device: Conveniently equipped with

encoders in order to capture the movements

performed with it.

Motor device: Fitted with servo-motors to

reproduce movement. In the design phase, this

device is also equipped with sensing elements in

order to carry out calibration and performance

measurements.

The present design is capable of completely valid

mobility. The sensing workbench can most

realistically imitate any type of operation. Both the

sensing and motor workbenches share a similar

design, so that they may or may not include motors

and sensors, that they may have the required

functionality on the basis of a single development.

The mechanical similarity between them likewise

eases the generation of movements from the stored

kinematics.

Various alternatives have been checked for

movement generation, among which we could

feature step-motors and servo-motors. The latter

presents suitable speed-torque characteristic curves,

which together with their simple handling have

made them the chosen solution. Different kinds of

movement must be generated, angular for the three

coordinate turns, and longitudinal for the cannula

and resection movements. In order to achieve linear

movements, mechanical transmission chains have

been designed, based on the turning of the servo-

motor.

The measurement of the three angular

movements is performed with angular encoders. The

linear movements are indirectly measured from the

electromechanical rotation system. In this way we

can use the same kind of sensor for the different

movements, with all the advantages of uniformity

and simplicity. Specifically, the encoder chosen will

be of the digital type and incremental.

Once the mechatronic system implemented, the

performance, both kinetic (speed, accelerations,

movement ranges and sensitivity) and dynamic

(torque) of the system was tested. For this,

automated tests have been devised to check the step

and ramp response of each motor device.

Finally, several tests were carried out by the

medical team, with the goal of simulating the

different kinds of movements that actual operations

might require. Two models were simulated:

cystoscopy (inspection) and resection (operation), as

well as a mixed model including various types of

movement. On the basis of these tests, we can state

that our system correctly replicates a model

operation.

3.3 Synchronization

In the system, various temporal distortions may

appear, which can generically be grouped as two

different sets of problems:

– Data delay: An ideal design in the FPGA will

make this negligible in the context of the time

intervals operated with.

– Loss of synchronization and regularity between

video frames and mechanical positions: Critical

aspect. The computers might be unable to send the

movement data in a totally regular and predictable

way while it reproduces a video or is handling other

processes.

Different alternatives have been checked:

synchronization by means of video subtitles, real-

time operative systems, and modulation in audio

channel.

The solution finally adopted is based on use of

the video’s audio channel. To include the digital

information of positions on an analog audio signal,

Manchester encoding has been chosen. It has been

decided to encode and send the sampling number

corresponding to each video frame, rather than all

the data for each frame. A Hamming code with

distance 4 is applied, to minimize environmental

noise. This strategy will make it possible to detect

up to three bit errors, and correct up to one bit. The

frame under the Hamming format is Manchester

encoded and modulated upon the audio channel.

Both processes have been carried out with Matlab.

The final application generates an audio file

containing the sampling numbers, spaced the exact

MECHATRONIC SYSTEM FOR TRANSURETHRAL RESECTION TRAINING

time needed. This file is included in the video, so

that we have a video with exact time marks

indicating a sampling number. In normal use, an

audio channel carries the frame number we are at,

and the other channel may include trainer’s

comments.

Before the use of a trainee’s workbench, a

massive download of positions in each motor system

is performed from the computer to the local FPGA.

During the video playing, the time marks are

extracted and the local memory is searched for the

associated positions.

3.4 Computer System

The computer system presents two layers. One of

them is opaque to a certain degree for the end user,

and collects the data of the different blocks and

integrates them. Also, it calculates the kinematics of

the resectoscope during the operation, using the

obtained data from the ultrasound location system.

Finally, it generates the movement references for the

device, synchronized with the endoscopic video.

The computer system has another aspect,

intended for the end user in the training context. A

simple and friendly user interface is offered for

interaction in the lecture room (Figure 5). It includes

a video player and a manager of the various

operation modes that the trainer may request. The

more tedious tasks, such as Bluetooth node

management and processes with data files have been

completely automated.

Figure 5: Computer system. User interface.

4 OPERATING THEATRE

POSITIONING

It is interesting to see how different data are

obtained in the operating theatre, such as the

endoscopic video of the operation and the actual in

situ trajectory followed by the resectoscope.

Obtaining the endoscopic video is possible with

commercial equipment, so we will concentrate on

the analysis of the problems regarding highly

accurate location. The different alternatives allowing

a sufficiently exact positioning were analyzed, with

the main focus on optical, radiofrequency and

ultrasound solutions. Several aspects, such as

economy and environmental constraints pointed to

ultrasound positioning as the ideal method.

Through high precision positioning by

ultrasound waves, it is possible to locate an object

within a given volume with a tiny error margin,

simply and quickly, without any necessary physical

contact with the point to be referenced (Fukuju,

2003; Casas, 2004; Mahajan, 2001; Prigge, 2000).

The initial idea was to obtain a system allowing

capture of the position of the resectoscope during an

actual operation in the operating theatre; the system

should therefore have very restrictive features.

In the case of our application, there are certain

key aspects defining the location system, due to the

characteristics of work in an operating theatre.

The range of the system must agree with the

dimensions of the operating theatre (in our case, up

to 3.5 metres).

Given the need for a precise reading of the

kinematics and of the position of the resectoscope,

the positioning refresh rate must be as high as

possible. It has been possible to obtain up to twenty

references per second.

In order to achieve millimetric errors in the

position of the resectoscope, it was necessary to

analyze other factors of the design: the possibility of

background noise, disadjustment of probes,

environmental factors (temperature and humidity),

and reflected ultrasound waves due to reflecting

surfaces.

The probes of the emitter modules (Figure 6) are

joined to the resectoscope in fixed positions. Three

probes are usually necessary in order to later infer

the position and direction of the instrument from

them.

The receiving probes will be attached to the

ceiling of the operating theatre in positions with

known coordinates, with the necessary precision to

later give references for the emitting probes in the

BIODEVICES 2008 - International Conference on Biomedical Electronics and Devices

space. A redundant number of probes make possible

to minimize the effects of occlusions.

The system includes the blocks shown in figure

The control system will send a signal to the

corresponding emitter modules, to generate the

ultrasound pulse train, and simultaneously the time-

of-flight of the ultrasound waves to the receiver

modules will be measured. During the time-of-flight

of the ultrasound pulses a measurement of the

temperature will be obtained, to compensate the

data. The features and pattern of the pulse trains

generated are critical for the system. Their

generation is based on a self-interference strategy, in

which the optimum phase and counter-phase periods

have been obtained analytically and empirically. In

the receiver, filtering is an equally delicate process,

articulated around a second order Rauch filter

followed by a high speed comparator.

The receiver modules will send the received

signal, filtered and conditioned, to the FPGA, which

will capture and process the data in order to

calculate the times-of-flight of each emitter-receiver

pair, as well as reliability indicators of each

measurement for their later processing. Once all of

the data have been processed, it will send all the

information obtained to the computer together with

the temperature measurement. The computer will

calculate the distances between each of the emitter-

receiver pair.

After obtaining the distances between emitter and

receiver probes, the coordinates of the emitter

probes are calculated by an algebra resolution

Figure 6: Emitter probe.

Figure 7: Precision positioning block diagram.

Figure 8: Positions of the end of the resectoscope.

Front Left

MECHATRONIC SYSTEM FOR TRANSURETHRAL RESECTION TRAINING

method (Casas, 2004) and a solution filtering

algorithm based on the least median of squares

(Casas, 2006). Once the positions of the emitter

probes have been obtained, the trajectory of the

resectoscope is obtained as well.

In order to verify the system, an experiment set-

up has been designed with the motor workbench

managed by the computer system in auto-calibration

mode. Figure 8 illustrates the results obtained. It

shows the position of an emitter probe located at the

end of the resectoscope. An emitter transducer is

attached to the mock up, which moves using a

predefined pattern. With three receivers, the results

are very accurate (few millimetres), and even the

bounces and oscillations of the mechanical system

itself can be detected.

5 CONCLUSIONS

Training surgery residents who start with endoscopic

operations remain a challenge. This paper has

described an electromechanical system developed

for learning the TUR technique. It consists of a

trainer’s workbench with a wireless sensing device

connected to several trainees’ workbenches with

motorised devices. These devices have a

resectoscope mounted on an electromechanical

structure able to reproduce all the movements of an

actual endoscopic operation.

The system has several operating modes that will

make it possible to:

– Reproduce the movements of an expert

surgeon in the hand of the trainee.

– Reproduce the pre-recorded movements of an

actual operation in the hand of the trainee.

– Assess the level reached by the student before

participation in any operations or in solving

problems requiring a certain degree of experience.

As a complement, a millimetrically accurate,

ultrasound-based positioning system has been

developed. This will be mounted on a resectoscope

in order to capture the movements performed in an

real operation. The management software of the

training room allows easy integration of these data

with the endoscopic video, to rely on an adequate

operations database.

What remains is to assess this tool in the

practical conditions of training urology residents in

the use of medical equipment, which will doubtless

offer most interesting data regarding the use or need

for modifications of the global system.

ACKNOWLEDGEMENTS

This work has been partially supported by the

Spanish Ministry of Science and Technology under

CICYT project numbers TIC2003-07766 and

TIN2006-15617-C03-02

REFERENCES

Anastasi, G., Bandelloni, R., Conti, M., Delmastro, F.,

Gregori, E. Mainetto, G., 2003. Experimenting an

Indoor Bluetooth-based Positioning Service.

ICDCSW’03, Proc. of the 23rd Int. Conf. on

Distributed Computing Systems Workshops, 480-483.

Ballaro, A., Briggs, T., Garcia-Montes, F., MacDonald,

D., Emberton, M., Mundy, A.R., 1999. A computer

generated interactive transurethral prostatic resection

simulator. The Journal of Urology, 162 (5), 1633-

1635.

Casas, R., 2004. BLUPS: Bluetooth and Ultrasounds

Positioning System. Doctoral Dissertation, University

of Zaragoza.

Casas, R., Marco, A., Guerrero, J.J., Falcó, J., 2006.

Robust Estimator for Non-Line-of-Sight Error

Mitigation in Indoor Localization. EURASIP J.

Applied Signal Processing, vol. 2006, article ID

43429, 1–8.

Fukuju, Y., Minami, M., Morikawa, H., Aoyama, T.,

2003. DOLPHIN: An Autonomous Indoor Positioning

System in Ubiquitous Computing Environment. Proc.

of the IEEE Workshop on Software Technologies for

Future Embedded Systems, 53-56.

Gettman, M.T., Hoznek, A, Salomon, L, Katz, R,

Borkowski, T, Antiphon, P, Lobontiu, A, Abbou, C.C.,

2003. Laparoscopic radical prostatectomy: Description

of the extraperitoneal approach using the Da Vinci

robotic system. The Journal of Urology, 170 (2 Pt 1),

416-419.

Gomes, M.P.S.F., Barret, A.R.W., Timoney, A.G., Davies,

B.L., 1999. A computer-assisted training/monitoring

system for TURP structure and design. IEEE Trans.

Information Technology in Biomedicine, 3 (4), 242-

251.

Katz, R, Nadu, A, Olsson, L.E., Hoznek, A, De La Taille,

A, Salomon, L, Abbou, C.E., 2003. A simplified 5-

step model for training laparoscopic urethrovesical

anastomosis. The Journal of Urology, 169 (6), 2041-

2044.

Kerfoot, B.P., Baker, H, Volkan, K, Church, P.A.,

Federman, D.D., Masser, B.A., De Wolf, W.E., 2004.

Development of validated instrument to measure

medical student learning in clinical urology: A step

toward evidence based education. The Journal of

Urology, 172 (1), 282-285.

Mahajan, A., Walworth, M., 2001. 3-D Position Sensing

Using the Differences in the Time-of-Flights from a

BIODEVICES 2008 - International Conference on Biomedical Electronics and Devices

Wave Source to Various Receivers. IEEE Trans.

Robotics and Automation, 17 (1), 91-94.

Ottensmeyer, M., Ben-Ur, E., Salisbury, K., 2000. Input

and Output for Surgical Simulation: Devices to

Measure Tissue Properties in vivo and a Haptic

Interface for Laparoscopic Simulators. MMVR'2000,

Proc. of Medicine Meets Virtual Reality, IOS Press,

236-242.

Prigge, E., How, J., 2000. An Indoor Absolute Positioning

System with No Line of Sight Restrictions a Building-

Wide Coverage. ICRA'00, Proc. IEEE Int. Conf. on

Robotics and Automation, 2, 1015-1022.

Pycha, A., Lodde, M., Lusuardi, L., Palermo, S.,

Signorello, D., Galantini, A., Mian, C., Hohenfellner,

R., 2003. Teaching transurethral resection of the

bladder: Still a challenge?. Urology, 62 (1), 46-48.

MECHATRONIC SYSTEM FOR TRANSURETHRAL RESECTION TRAINING