Sensors and Actuators A 162 (2010) 291–296
Contents lists available at ScienceDirect
Sensors and Actuators A: Physical
journa l homepage: www.e lsev ier .co
Miniat abl
Daniele C ca C
Arianna
a NEURICAM s.r
b CRIM Lab, Scu
a r t i c l
Article history:
Received 30 Se
Received in re
Accepted 17 M
Available onlin
Keywords:
Camera modu
Disposable en
Endoscopic ca
for
ped.
ed o
com
imag
and t
fers t
nd d
1. Introdu
Standard
adopted since few years ago and still widely diffused today are
based on the use of a flexible endoscope which includes a light
delivery system, an imaging camera and some additional channels
for medical purposes. Light from an external source is carried to
the target t
tissue are t
fibres or a le
body. Imag
Commercia
a length of
till 13mm a
inserted int
assess the i
tively. Gene
the patient
the instrum
other featu
thesia (e.g.
required fo
Due to th
dard gastro
as Crohn’s
∗ Correspon
E-mail add
kind
ent
a lon
as push enteroscope, is introduced, into the upper gastrointestinal
tract to examine and evaluate the proximal section of the small
bowel [3]. Double balloon enteroscopy is a newly developed endo-
scopic method allowing exploration of the small intestine in steps
0924-4247/$ –
doi:10.1016/j.hrough an optical fiber and the images of the observed
ransferred back through a bundle of coherent optical
ns system to an acquisition camera located outside the
e is thus displayed on a video for diagnostic purposes.
l video gastroscopes have a diameter of 9–12mm and
110 cm, while video colonoscope has a high diameter
nd a length from 70 to 160 cm. These instruments are
o the human body through the mouth or the anus to
nterior surfaces of the stomach and the colon, respec-
rally these medical procedures are carried out while
is awake. Despite the reduction in the diameter of
ent and the improvement in terms of flexibility and
res, the procedure is still painful, therefore local anes-
for gastroscopy) or even sedation (for colonoscopy) are
r some patients [1,2].
e difficulty to reach the small bowel by mean of a stan-
scope, pathologies of this gastrointestinal tract, such
disease or obscure gastrointestinal bleeding, need dif-
ding author. Tel.: +39 050 883405; fax: +39 050 883497.
ress: c.cavallotti@sssup.it (C. Cavallotti).
by using two balloons to grip the intestinalwall; the endoscope can
be inserted for 430 cm inside the small bowel [4]. These procedures
are invasive, cause high discomfort to the patient and need to be
performed under general anesthesia.
The advent of video capsule endoscopy (VCE) in year 2000 has
dramatically changed the diagnosis and management of many dis-
easeof the small intestine, turning inspectionof thegastrointestinal
tract into non-invasive and almost completely painless examina-
tion [5,6].
1.1. The endoscopic capsule
The endoscopic capsule (EC) is a small device with the size and
shape of an antibiotic pill which is easily swallowed by the patient
and transmits the images during its transit through the gastroin-
testinal tract. These features permit, for example, to investigate the
entire small bowel with benefits in terms of patient comfort and
reduced risk of side effects. Themain building elements of an EC are
an imaging sensor with lens and light-source (the camera module),
a battery, a wireless module for data exchange and some glue elec-
tronics. All these parts are contained in a bio-compatible enclosure
(with a transparent window), suited to withstand the conditions of
see front matter © 2010 Elsevier B.V. All rights reserved.
sna.2010.03.031urized digital camera system for dispos
ovia, Carmela Cavallottib,∗, Monica Vatteronib, Lu
Menciassib, Paolo Dariob, Alvise Sartori a
.l., Trento 38100, Italy
ola Superiore Sant’Anna, Pisa 56100, Italy
e i n f o
ptember 2009
vised form 12 March 2010
arch 2010
e 30 March 2010
le
doscopy
psule
a b s t r a c t
A miniaturized color camera module
surgery has been designed and develo
conductor (CMOS) sensor, miniaturiz
connector on a single substrate. The
mination, VGA resolution and good
demonstration system has been built
cable to a receiver board, which trans
ware controls the hardware setting a
image processing tasks.
ction
screening procedures of the gastrointestinal tract
ferent
balloon
wherem/locate /sna
e endoscopic applications
lementela, Pietro Valdastrib,
disposable endoscopic applications and minimally invasive
The module consists of a Complementary Metal Oxide semi-
ptics, a Light Emitting Diode (LED)-based illuminator and a
pact size (5.0mm×8.2mm×7.0mm), high-efficiency illu-
e quality allow it to be used in endoluminal procedures. A
ested in vivo. The module is connected through a 1.5-m long
he data stream to a Personal Computer (PC). A dedicated soft-
isplays the image, after having performed various color and
© 2010 Elsevier B.V. All rights reserved.
s of examination, named push enteroscopy or double
eroscopy. Push enteroscopy is an invasive procedure
ger and more rigid gastrointestinal endoscope, known
292 D. Covi et al. / Sensors and Actuators A 162 (2010) 291–296
the bowels. Beside of the basic functionality, some advanced fea-
tures have been recently introduced. It is worth to mention a drug
reservoir for targeted delivery of an active principle, pH and tem-
perature sensors for monitoring of relevant parameters, actuators
enabling active locomotion [7,8].
Despite of these innovations, the camera module remains the
heart of the EC. Its design specifications are partially influenced
by the biological environment which the capsule is targeted to.
For example, the examination of the oesophagus is very fast and
requires a high frame rate and a wide field of view, while the tran-
sit through the intestine lasts for many hours and an accordingly
long lasting battery is necessary [9,10]. The requirements of low
power consumption, good image quality and small size have to be
met for any application. Low cost of the module is mandatory in
case of a disposable application such as an endoscopic capsule.
Considering these purposes, a camera module based on a com-
mercial complementary metal oxide semiconductor (CMOS) color
imager was developed and any component or technology used was
selected on the basis of its contribution to the achievement of the
mentioned requirements. A demo system was set up to evaluate
a preliminary wired prototype of the module, which was success-
fully tested during ex vivo and in vivo experiments on a porcine
model.
2. System overview
2.1. Camera module
The camera module prototype (Fig. 1(a)) is based on a commer-
cial CMOS color imager. A device already available on the market
was chosen to reduce development costs. VGA image resolution
(640×480
of 1.1mm×
data bus ac
imaging arc
signal integ
of impedan
lines. Amon
Fig. 1. The fir
cent coin (a). A
Table 1
Features of image sensor.
Parameter Value
Optical format 1/11-in. VGA (4:3)
Active area size 1.43mm×1.07mm
Die size 2.46mm×2.73mm
Active pixels count 648×488
Pixel size 2.2m×2.2m
Color filter array RGB Bayer pattern
Frame rate Programmable up to 30 fps
Responsivity 1.1V/lux s
Dynamic range 64dB
Signal to noise ratio (max) > 36.5dB
Power consumption 80mW
Table 2
Features of ca
Parameter
Dimension
Power consu
Number of c
Data output
Image data o
Control data
Master clock
Field of view
Lens f/#
LED luminou
was selecte
this approa
substrate a
Three high-
0603
d len
of 60
, sys
-sp
vera
pow
eatur
ma
mod
on a
dule is dominated by the connector. The used component
lected for an easy connection of the prototype with a 0.5mm
at flexible PCB where a 24MHz oscillator and the 12 wire of
le are soldered (Fig. 1(b)).
rder to better quantify the contribution of the presented
a module, it is meaningful to compare its performance with
LCAM SB2, which can be considered the gold standard for
e endoscopy (Table 3). The lightning system of the cam-
dule is made up of 3 high-efficiency LEDs soldered to the
CB where the imaging sensor is attached on. Because of a
l design, no shadows or illumination non-uniformities are
by the asymmetrical layout around the optical axis and the
position relative to the lens top. These two features allow
e of a single board, lowering the cost and simplifying the
rison between the proposed camera module and PILLCAM SB2.
Camera module PILLCAM SB2
l dimension (diameter) 10.5mm 11mm
l system 1 lens 3 lenses
f view 60◦ 156◦
tion 640×480 256×256
er of LED 3 6pixels) with 2.2m pixel size results in an active area
1.4mm in size. Pin count is limited by the use of serial
cording to I2C protocol on input and standard mobile
hitecture (SMIA) protocol on output [11]. Long-range
rity for the image stream is guaranteed by the use
ce-matched low-voltage differential signaling (LVDS)
g the various packaging options available, the die form
st prototype of the camera module, with its size compared to a D 1
detail of camera cable (b).
print (
molde
(FOV)
supply
and low
lines. O
at full
main f
are sum
The
minati
the mo
was se
pitch fl
the cab
In o
camer
the PIL
capsul
era mo
same P
carefu
caused
lower
the us
Table 3
A compa
Fronta
Optica
Field o
Resolu
Numbmera module.
Value
5.0mm×8.2mm×7.0mm
mption 190mW max
onnection 12
rate 140Mbps
utput protocol SMIA LVDS
input protocol I2C
16–24MHz
60◦
2.8
s efficiency 46 lm/W
d to adopt a chip-on-board assembly technology. In
ch the die is attached on a printed circuit board (PCB)
nd its signal pads are wire-bonded on the PCB tracks.
efficiency (46 lm/W) white LEDs with small PCB foot-
) provide the proper amount of light. A low-cost plastic
s focus the image on the sensor with a field of view
◦ (diagonal). Electrical connectivity is limited to power
tem clock, high-speed image transfer (140Mbps LVDS)
eed system control (I2C bus), with an overall count of 12
ll power consumption is limited to 190mW with LEDs
er and imager working at 30 frames per second. The
es of both the imaging sensor and the camera module
rized in Tables 1 and 2.
ule is 5.0mm×8.2mm×7.0mm, including optics, illu-
nd connector. It is worth to observe that the height of
D. Covi et al. / Sensors and Actuators A 162 (2010) 291–296 293
Fig. 2. Hardware architecture of the demonstration system.
assembly process. The camera module has overall dimensions of
5.0mm (w)×8.2mm (h) which fits in a capsule with a minimum
innerdiameterof 9.6mm.Thanks to its rectangular shape, anempty
space of about 29mm2 is left free for future integration of specific
tools, such as biopsy needles.
The optical systemof the PILLCAMSB2 has a diameter of 11mm.
It ismade u
ture stopan
a transpare
ments lead
presented c
which focus
of 60◦ (diag
module dur
methacryla
to protect t
be coupled
negative len
and inner s
FOV of 156◦
2.2. Demon
A demo
formance o
conditions
cameramod
eter of 4mm
data stream
Virtex5 field
into adoubledata rate synchronousdynamic randomaccess
ry /DDR2 SDRAM) used as a video buffer. The raw image is
rred to a host computer using a USB 2.0 host–client inter-
hrough this link, the computer sends to the board the values
ritten into imager registers, the power level to be set for
D driver and other low-level settings. A specifically devel-
oftware (Fig. 4) runs on the host PC, allowing the user to
l the main high-level hardware settings (integration time,
hannels gain, LED power, etc.). The software handles also
ta stream from the camera module and performs a series of
processing tasks on the received raw data before displayingp of 3 lenses (1 focusing lens and 2 field lenses), an aper-
d the illumination sources,which arepositionedbehind
nt elongated dome [12]. The combination of these ele-
s to a 156◦ (diagonal) FOV. The optical system of the
ameramodule is composed by a short-focal-length lens
es the light ray onto the optical sensor, achieving a FOV
onal). In order to preliminary evaluate the presented
ing ex vivo and in vivo tests, a transparent Poly-methyl
te (PMMA) window was placed in front of the system
he camera from organic material. However, a dome to
with the present lens is under design. It will acts as a
s by using a different radius of curvature for the outer
urface. This dome will enable the system to achieve a
(diagonal).
stration system
nstration system (Fig. 2) was built to verify the per-
f the camera module in the realistic environmental
of in vivo experiments. For these preliminary tests, the
ulewaswired througha1.5-m long cable (with adiam-
) to a custom receiver board (Fig. 3). The serial image
is decoded by a SMIA receiver interfaced to a Xilinx
programmable gate array (FPGA). The FPGA stores the
stream
memo
transfe
face. T
to be w
the LE
oped s
contro
color c
the da
imageFig. 4. Software architecture of the demonstraFig. 3. Demonstration system.tion system.
294 D. Covi et al. / Sensors and Actuators A 162 (2010) 291–296
Fig. 5
them on vid
effectivenes
of their num
• Bayer pat
missing d
matrix (d
gradient-
along a s
slightly in
tions, but
details.
• Lens vign
nonunifor
image corners is removed by rescaling each pixel value using cor-
rection coefficients. The coefficients are lens-depending and are
obtainedwithacalibrationprocedureona reference image, based
on a best-fit procedure with a two-dimensional second-order
polynomial model.
• Bad pixel removal. Removal of the so-called ‘hot pixels’, which are
pixels appearing white regardless of light conditions. This mal-
functioning isdue to tolerances in thechipmanufacturingprocess
and it is corrected by means of a median spatial filtering based
on a 3×3 pixel kernel.
• White balance. A procedure consisting in a global adjustment of
the intensities of the colors so that objects which appear white in
the scene are rendered white in the image. The adjustment is car-
ried out using a 9-coefficient matrix obtained from a calibration
procedure with a 24-patch MacBeth ColorChecker target. Coeffi-
cients are calculated with a white-point-preserving least-square
regression routine, which minimizes the difference between the
acquired RGB coordinates of the patches and their reference val-
ues. White point is then adjusted on the basis of the knowledge
of the LED chromatic coordinates in the CIE1931 color space
[13].
• Edge enhancement. This step isbasedona spatial filterwithaSobel
3×3 pixel kernel, which enhances edge transitions. The main
effect of this filtering is to make details to appear sharper. Ideally,
rm regions are left unchanged but a common drawback is
ncrease of their noise level. This unwanted effect is avoided
e implemented version, where the filtering is applied only
re an edge has been previously detected [14].
ma correction. Image data are converted to the standard RGB
space, used by the Microsoft Windows operating system
ive the monitor. This step includes the application of the
ma shift, which takes into account the non-linear sensitivity
e human eye to variation of luminance levels.. Acquired images during ex vivo tests in porcine stomach.
eo. The quality of the displayed image depends on the
s of the implemented algorithms and on a fine tuning
erical parameters. The main processing tasks are:
tern demosaicing. Restoration of the color information
ue to the Bayer color filter pattern applied on the pixel
emosaicing). The use of a first neighboring interpolating
unifo
the i
in th
whe
• Gam
color
to dr
gam
of thsensitive algorithm reduces artifacts which can appear
harp edge in the image. Computational complexity is
creased, if compared to other simple interpolating solu-
leads to considerably better color rendering of tissue
etting correction. Compensation of the illumination
mity caused by lens vignetting. Light shading at the
Fig. 6. In vivo experiments of the camera module. Fig. 7. Acquired images during in vivo tests.
D. Covi et al. / Sensors and Actuators A 162 (2010) 291–296 295
The described flux guarantee a good rendering of the finest
image details and, together with a calibrated monitor, an accu-
rate color matching between the observed tissue and the displayed
image. The version developed for the demonstration system is
implemented in C++ language. The computational complexity of
the algorithms and the high-level programming language imply
the use of a high-performance PC to obtain images streaming at 30
frames per second.
3. Experimental results
Image quality was tested in ex vivo trials, in order to tune the
hardware and software setting of the system. Some freshly excised
porcine gastric tissues were used to acquire images (Fig. 5). Once
the correct parameter settings were defined, in vivo experiments
were performed on a porcine model. The aim of these trials was to
test the performance of the cameramodule by evaluating the effec-
tiveness of illumination as well as the image quality in a realistic
environment.
The experiments were carried out in a specialized experimental
animal facility, with the assistance and collaboration of a specially
trained medical team in compliance with the regulatory issues
related to animal experiments. The camera module was located
in a capsular shell with size of 13mm (diameter)×20mm (length)
manufacturedwith rapid prototyping printing, while a PMMA slide
was placed in front of the camera module. In order to avoid misting
Fig. 8. Acq
up due to temperature gradient the slidewas coatedwith a anti-fog
coating. Additionally a hydrophobic coatingwas sprayed toprevent
the deposit of organic material.
The capsule was introduced in the stomach through a 5mm
laparotomic incision (Fig. 6) and manually oriented to visualize the
most interesting regions of the gastric cavity. The light level was
regulated by software in order to achieve the best quality of the
image displayed on the PC. A video was recorded for the average
duration of a detailed gastroscopy (30min), and two of its frames
are shown in Fig. 7. Another test was performed introducing the
capsule in the colon, through the anus, in order to reproduce a
colonoscopy. The analysis of the acquired movie indicates that the
proposed camera module is able to evenly illuminate the inner tis-
sue,with a light level suitable to obtain low-noise images. The good
focusing and color rendition allow the features of the observed
target to be properly displayed, thus enabling the endoscopist to
reliably perform a diagnosis.
A comparison on two images acquired by the camera module
and the PILLCAM SB2 during in vivo tests has been carried out. The
amount of light provided by the designed illumination system is
comparable with a PILLCAM SB2 image taken in similar conditions
(Fig. 8). From an illumination viewpoint, images obtained with the
PILLCAM SB2 and the presented module are comparable, but with
half the number of LEDs, half the footprint and, with a reasonable
confidence, half the power consumption. This result is obtained
through a careful selection of high-efficiency LEDs and a high sen-
sitivity camera. Sinceno imageswith reference targets are available
for the PILLCAM SB2, a comparison on two images acquired during
in vivo tests has been carried out. A 40×40 pixels region with-
out textures or features related to the observed target and with
an even illumination was identified and selected in each image.
The histogram of the pixel code distribution was calculated (Fig. 9).
As resulting from the pictures, the selected regions have the same
istog
moduuired image from PILLCAM SB2 (a) and from camera module (b).
Fig. 9. H
cameraram of PILLCAM SB2 image selected region (a) and of the proposed
le selected region (b).
296 D. Covi et al. / Sensors and Actuators A 162 (2010) 291–296
mean value even if the PILLCAM SB2 illumination system is com-
posed by 6 LEDs while the presented camera module has only 3
LEDs. Assuming that the two targets have a comparable reflectiv-
ity, this result could be achieved by the use of different integration
times in the
of the two
the PILLCAM
the flat regi
ule. This me
module.
4. Conclus
The sma
swallowabl
for battery-
terms of res
physicians t
of the chos
to be used
fully tested
model.
A new v
opment. It w
be tailored
rithm is goi
implement
be therefor
used to disp
Acknowled
The wor
Commissio
EU/IST-200
and all the t
References
[1] D.K. Rex,
of a shap
terology
[2] A. Eickho
ert, D. Sc
endoscop
American
[3] L.F.Musc
of push e
[4] H. Yamam
ano, Tota
Gastroint
[5] P. Swain,
terol 14 (
[6] M. Water
Imaging 3
[7] C. McCaff
technolog
[8] P. Valdas
new mec
ronments
[9] B. Koslow
improved
Endoscop
[10] I.F. Urien
World Jou
[11] http://ww
[12] In-vivo im
[13] G.D. Fina
258–261,
querque,
[14] R.C. Gonz
Biographies
Daniele Covi graduated in physics (summacum laude) from theUniversity of Trento
(Italy) in 2001 working on the active control of magnetic fields for atomic traps. In
receiv
e join
of CM
senso
rrently
desig
d min
Cava
from
e is a
a in P
Vatte
ical en
from t
at Ne
ible fo
cuola
ible fo
cal ap
public
-nois
.
ment
ty of
. He
tures
tration
He is
nt vis
aldas
the
b of t
his P
titled
w assi
ntabl
urop
cal de
Menc
ty of P
e San
on th
e rece
for M
profe
n rese
ricati
orking
of mic
ario re
isa in
e San
e Scho
ole Po
da Un
e San
. His
ndmi
ions. H
o boo
He is
. He i
embe
and a
n Rob
tion So
of th
artori
d a Ph
e Cen
g of fl
he join
SI Des
com
ems fotwo imaging sensors andbyadjusting theoutputpower
lightning system. Moreover, the standard deviation of
SB2 histogram is higher than the one obtained from
on of the image taken with the presented camera mod-
ans a lower fixed pattern noise (FPN) for the presented
ions and future work
ll size achieved makes the module easily fitting into a
e capsule and the overall power consumption allows
supplied operation. The quality of acquired images in
olution and color rendition is good, as required by the
o perform correct and reliable diagnosis, while low cost
en components and technologies allows the module
in disposable applications. The prototype was success-
during ex vivo and in vivo experiments on a porcine
ersion of the camera module is currently under devel-
ill have a smaller size and a wireless interface and will
on an endoscopic capsule. The image processing algo-
ng to be ported from C++ language to VHDL code for an
ation on FPGA device. The PC computational load will
e greatly reduced allowing lower-end computers to be
lay the image stream.
gments
k described in this paper was funded by the European
n in the framework of VECTOR FP6 European project
6-033970. The authors are also grateful to Dr. Burchielli
eam for the help during the testing phase of the device.
M. Khashab, G.S. Raju, J. Pasricha, R. Kozarek, Insertability and safety
e-locking device for colonoscopy, The American Journal of Gastroen-
100 (2005) 817–820.
ff, J.V. Dam, R. Jakobs, V. Kudis, D. Hartmann, U. Damian, U. Weick-
hilling, J.F. Riemann, Computer-assisted colonoscopy (the neoguide
y system): results of the first human clinical trial (“pace study”), The
Journal of Gastroenterology 102 (2) (2007) 261–266.
arella, Endoscopic shuffling, infection control, and the clinical practice
nteroscopy, Gastroenterology Nursing 30 (2) (2007) 109–115.
oto, Y. Sekine, Y. Sato, T. Higashizawa, T.Miyata, S. Iino, K. Ido, K. Sug-
l enteroscopy with a nonsurgical steerable double-balloon method,
estinal Endoscopy 53 (2) (2001) 216–220.
The future of wireless capsule endoscopy, World Journal of Gastroen-
26) (2008) 4142–4145.
man, R. Eliakim, Capsule enteroscopy of the small intestine, Abdom
4 (4) (2009) 452–458.
rey, O. Chevalerias, C. O’Mathuna, K. Twomey, Swallowable-capsule
y, IEEE Pervasive Computing 7 (1) (2008) 23–29.
tri, R.J. Webster, C. Quaglia, M. Quirini, A. Menciassi, P. Dario III, A
hanism for mesoscale legged locomotion in compliant tubular envi-
, IEEE Transactions on Robotics 25 (5) (2009) 1047–1057.
sky, H. Jacob, R. Eliakim, S. Nadler, Pillcam eso in esophageal studies:
diagnostic yield of 14 frames per second (fps) compared with 4 fps,
y 38 (1) (2006) 27–30.
, C. Carretero, A. Borda, M. Muoz-Navas, Colon capsule endoscopy,
rnal of Gastroenterol 14 (34) (2008) 5265–5268.
w.smia-forum.org.
aging device and optical system thereof, WO2007060659 (2007).
lyson, M.S. Drew, White-point preserving color correction, pp.
in: Proceedings of the IS & T/SID 5th Color Imaging Conference, Albu-
New Mexico, 1997.
alez, R.E. Woods, Digital Image Processing, Prentice Hall, 2007.
2005 he
(Italy). H
duction
imaging
He is cu
systems
tems an
Carmela
Honors)
rently sh
Sant’Ann
Monica
in electr
Physics
worked
respons
ing at S
respons
biomedi
journal
sors, low
systems
Luca Cle
Universi
in FPGA
architec
demons
rithms.
intellige
Pietro V
ors) from
CRIM La
obtained
a thesis
He is no
of impla
several E
biomedi
Arianna
Universi
Superior
program
1999, sh
Grippers
she is a
Her mai
microfab
She isw
opment
Paolo D
sity of P
Superior
ics at th
at the Ec
at Wase
Superior
students
tronics a
applicat
tor of tw
papers.
Research
tion, a m
Systems
actions o
Automa
Robotics
Alvise S
1978 an
joined th
modellin
In 1990
of the VL
Trento, a
and systed the MBA from the Alma Graduate School – University of Bologna
ed Neuricam in 2000 participating in the design and transfer to pro-
OS optical sensors and setting up the Electro-Optical Laboratory for
rs characterization. He was head of the VLSI Design Area since 2002.
working as project manager in the field of advanced electro-optical
n. His research interests focus on optical distance measurement sys-
iaturized camera modules for endoscopy applications.
llotti received her Laurea Degree in Biomedical Engineering (with
the Campus Bio-Medico University in Rome in December 2007. Cur-
Ph.D. student in Biorobotics at the CRIM Lab of the Scuola Superiore
isa.
roni was born in La Spezia, IT, in 1975. She received the M.S. degree
gineering from University of Pisa, Pisa, IT, in 2001 and PhD degree in
he University of Trento, Trento, IT, in 2008. From 2002 to 2008, she
uriCam, Trento, IT, as Pixel Engineer and analog designer, becoming
r CMOS Image Sensor development in 2005. Presently she is work-
Superiore Sant’Anna, Pisa, IT, as post-doctoral fellow where she is
r research and development of image sensors and vision systems for
plications. She is the author or coauthor of a few conference and
ations and three patents. Her interests include CMOS image sen-
e analog electronics, high dynamic range pixels and endoscopic vision
el received the B.S. degree in communication engineering from the
Trento in 2001 developing a digital neural network implemented
joined Neuricam Srl, Trento, in 2001, where he designed digital
in programmable logic devices for vision systems like glue logic for
baseboards of optical sensors and complex image processing algo-
currently an HDL developer and a project manager in the field of
ion systems design.
tri received his Laurea Degree in Electronic Engineering (with Hon-
University of Pisa in February 2002. In the same year he joined the
he Scuola Superiore Sant’Anna in Pisa as Ph.D. student. In 2006 he
h.D. in Bioengineering from Scuola Superiore Sant’Anna by discussing
“Multi-Axial Force Sensing in Minimally Invasive Robotic Surgery”.
stant professor at CRIM Lab, with main research interests in the field
e robotic systems and active capsular endoscopy. He is working on
ean projects for the development of minimally invasive and wireless
vices.
iassi received her Laurea Degree in Physics (with Honours) from the
isa in 1995. In the same year, she joined the CRIM Lab of the Scuola
t’Anna in Pisa as a Ph.D. student in Bioengineering with a research
e micromanipulation of mechanical and biological micro objects. In
ived her Ph.D. degree by discussing a thesis titled “Microfabricated
icromanipulation of Biological and Mechanical Objects”. Currently
ssor of biomedical robotics at the Scuola Superiore Sant’Anna, Pisa.
arch interests are in the fields of biomedical micro and nano-robotics,
on technologies, micromechatronics and microsystem technologies.
on several Europeanprojects and international projects for the devel-
ro and nano-robotic systems for medical applications.
ceived his Laurea Degree inMechanical Engineering from theUniver-
1977. Currently, he is a professor of biomedical robotics at the Scuola
t’Anna, Pisa. He also established and teaches the course onMechatron-
ol of Engineering, University of Pisa. He has been a visiting professor
lytechnique Federale de Lausanne (EPFL), Lausanne, Switzerland, and
iversity, Tokyo, Japan. He is the director of the CRIM Lab of Scuola
t’Anna, where he supervises a team of about 70 researchers and Ph.D.
main research interests are in the fields of medical robotics, mecha-
croengineering, and specifically in sensors and actuators for the above
e is the coordinator of many national and European projects, the edi-
ks on the subject of robotics and the author of more than 200 journal
a member of the Board of the International Foundation of Robotics
s an associate editor of the IEEE Transactions on Robotics andAutoma-
r of the Steering Committee of the Journal of Microelectromechanical
guest editor of the Special Issue onMedical Robotics of the IEEE Trans-
otics and Automation. He serves as president of the IEEE Robotics and
ciety and as the co-chairman of the Technical Committee on Medical
e same society.
received an M.A. degree in Physics from the University of Oxford in
. D. in Geophysics from Imperial College, London, in 1983. He then
tral Research Laboratory of Olivetti, where he carried out research on
uido-dynamic systems and design of digital CMOS integrated circuits.
ed IRST, a Research Institute in Trento, Italy, where he was in charge
ign Laboratory. Since 1998, he is President and CEO of NeuriCam SpA,
pany he co-founded in 1998, active in the fabless production of chips
r computer vision.