- Email: [email protected]
Piezoelectric tactile sensor for submucosal tumor detection in endoscopy
G Model
ARTICLE IN PRESS
SNA-9604; No. of Pages 11
Sensors and Actuators A xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Sensors and Actuators A: Physical journal homepage: www.elsevier.com/locate/sna
Piezoelectric tactile sensor for submucosal tumor detection in endoscopy Cheng-Hsin Chuang a,∗ , Tsan-Hsiu Li a , I-Chinms Chou b , Ying-Juimr Teng b a b
Department of Mechanical Engineering, Southern Taiwan University of Science and Technology, Tainan 71005, Taiwan Kuang Tai Metal Industrial Co. Ltd., Kaohsiung, Taiwan
a r t i c l e
i n f o
Article history: Received 30 September 2015 Received in revised form 6 April 2016 Accepted 6 April 2016 Available online xxx Keywords: Tactile sensor Piezoelectric Endoscopy Stiffness
a b s t r a c t In this study, we have fabricated a miniaturized piezoelectric tactile sensor (Ø = 1.4 mm) that is suitable for mounting on an endoscope to detect submucosal tumors. The sensing mechanism is based on a tandem spring model and consists of two components with varying stiffness, namely a hard inner structure embedded in a soft outer packaging. The voltage output of the PVDF sensing film is proportional to the localized normal stress exerted by each component and this differential output can be used to extract information about the elasticity of the test object/biological tissue. The sensor design has been tested by embedding different elastomers and artificial tumors in a pig’s stomach for simulating the conditions of submucosal tumors in humans and shows good agreement with theoretical analysis and numerical simulations. The sensor response is proportional to the Young’s modulus of test sample over a range of 1.01–3.51 MPa, making it suitable for detection of tumors present in soft tissues. The proposed miniaturized tactile sensor utilizes a passive sensing element, is wired to the external readout system through the metal catheter which can be inserted into the endoscopic channel, is safe for insertion into the stomach as it has a biocompatible packaging and can thus be utilized to provide tactile information during regular endoscopy. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Cancer typically manifests as hard lumps (tumors) in the soft tissues of the stomach, breast, prostate, lungs and other body organs [1]. Many studies have also demonstrated that tumors are significantly stiffer than the surrounding tissue [2–4]. Submucosal tumors (SMT) in clinical terminology are protuberant lesions or bumps covered with intact mucosa and their etiology varies from non neo-plastic lesions to true neoplasia [5,6]. SMT of the stomach are difficult to diagnose at an early stage using luminal endoscopy or barium radiography as these lesions originate in the muscularis mucosa or submucosa instead of the surface of gastric mucosa and visual detection may not be sufficient. Usually, endoscopic ultrasonography (EUS) can easily show whether a submucosal lesion is a tumor and is considered the most accurate imaging technique for the diagnosis of gastrointestinal SMT [7,8]. However, the facility of EUS is not usually available in low resource settings due to
∗ Corresponding author. E-mail address: [email protected] (C.-H. Chuang).
its high set up cost. We propose an alternative way to detect SMT based on direct contact with a miniaturized tactile sensor that is compatible with traditional endoscopy for the detection of suspicious lesions in the upper gastrointestinal tract. Tactile sensors are devices that can measure certain properties of an object or contact event through physical contact between the sensor and the object [9] and can be categorized based on the transduction mechanism that they utilize such as capacitive sensing [10], piezoelectricity [11], piezo resistivity [12] and pneumatic-based methods [13] to name a few. During MIS, much of the tactile information that is normally available during open surgery is lost. Real time measurement of elasticity or stiffness of biological tissues would help the surgeon to locate blood vessels, determine the health and type of a tissue or identify cancerous tumors whose mechanical properties are known to differ substantially from healthy tissue. Recently, many researchers have presented different medical tactile sensors with a variety of designs and principles to measure the tactile information during MIS. Sokhanva et al. [14] proposed a piezoelectric tactile sensor (size: 22 × 4 × 0.8 mm) for tissue characterization by attaching three PVDF films to two stiff supporting beams and a flexible beam to acquire the voltage output as a mea-
http://dx.doi.org/10.1016/j.sna.2016.04.020 0924-4247/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS
2
C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
sure of compliance of the object in contact. Tanaka et al. [15] proposed a real time tactile sensor system (size: Ø = 4 mm) based on balloon expansion by injection of biocompatible water and can ensure sterilization of the fluid with safe contact while requiring no electrical power. The sensor output results can effectively study both the stiffness and surface condition with potential for intraoperative brain tumor diagnosis. Sangpradit et al. [16] proposed a force feedback sensor (size: Ø = 11.5 mm) for use in MIS. Its distinguishing characteristic as compared to other tactile and force sensors is the air cushion on which the main sensing element is located. Zhao et al. [17] used the two-spring model to fabricate a tactile sensor (size: 19 × 19 mm) utilizing two piezo resistive cantilevers of varying stiffness to detect the hardness of biological tissues. Our previous work employed a tactile sensor (size: 6.5 × 3.5 mm) that can be attached to a laparoscope for differentiating soft tissues in animals [18]. The mechanical characteristics of the contact object can be differentiated by extracting the voltage outputs from the piezoelectric film at the inner and outer electrodes. Åstrand et al. [19] developed a resonance based piezoelectric tactile sensor (size: 15 mm × 5 mm) that can detect stiffness variations located up to 4 mm in silicone and chicken muscle, mimicking the presence of tumors embedded in prostate tissue. Using this tactile sensor, the shift of the resonance frequency and the force at contact with tissue can be measured and combined into a tissue stiffness parameter. Kwon et al. [20] proposed a simple tactile sensor (size: 26 mm × 8 mm) that has multiple sensing points using a semiconductor micro strain gage for recognition of tissue stiffness. The sensor exhibits high linearity with comparable spatial resolution of a human finger and can be attached to the gripper of surgical instruments for in-vivo palpation during MIS. While the aforementioned concepts can aid in improving tactile sensing during MIS, they are limited by their size and design for practical application in conjugation with regular endoscopy for obtaining information about tissue elasticity and detection of submucosal tumors (endoscopic channel Ø = 2.3 mm). An endoscope is a useful MIS tool as it not only provides an image for visual inspection but also enables biopsies and the retrieval of foreign objects from inner organs of the body that are difficult to access. However, a visual inspection for an early diagnosis of a tumor is still often insufficient and thus obtaining tactile feedback using an endoscope would be helpful for increasing the precision of examinations and thus improving the quality of treatments. In this study, we have developed a miniaturized tactile sensor that can be mounted on an endoscope and can enable the surgeon to locate and differentiate the SMT from normal tissues in the stomach. We have tested the efficacy of the proposed tactile sensor for elasticity detection of commercial elastomers and artificial tumors embedded in the submucosa of a pig’s stomach. Furthermore, the theoretical analysis in ideal contact conditions based on the tandem spring model shows good agreement with observed simulation and experimental results and can operate in a dynamic measurement range suitable for identifying tumors in soft biological tissues.
2. Sensor design 2.1. Sensing mechanism The miniaturized tactile sensor consists of two components of varying stiffness, namely a hard inner component (E1 ) consisting of a copper ball and a soft outer packaging (E2 ) made of PDMS. When the sensor contacts a test object under a uniform force, the two components undergo varying deformation depending on their stiffness in comparison to the stiffness of the test object. Since the Young’s modulus of the inner structure is greater than that of the outer packaging material (E1 > E2 ), a non-uniform stress distribu-
tion will be experienced by the piezoelectric film, resulting in two varying output voltages obtained at the corresponding structural electrodes. Since the force transfer by the two sensor components of varying stiffness is different under a uniform applied load, the charge generated (voltage output) by the piezoelectric film under the two components will also be different. When the tactile sensor contacts soft objects, the deformation of the outer packaging will be significantly greater than when the sensor contacts hard objects. Thus the force transfer by the outer structure will be relatively larger when the sensor contacts soft tissues as compared to hard objects. Consequently, there is a lower stress distribution differentiation on the piezoelectric film corresponding to the two sensor components when contacting a soft object and the ratio of the output voltages (V1 /V2 ) is relatively small as shown in Fig. 1(a). However, when contacting hard objects, the majority of the normal force will be transferred by the inner structure, resulting in a greater stress distribution differentiation and so the ratio of output voltages is greater, as shown in Fig. 1(b). Therefore, the ratio of output voltages from the inner structure and outer packaging material can be used to obtain information about the hardness (and corresponding stiffness or elasticity) of the test object. 2.2. Ideal contact analysis using the tandem spring model The tandem spring model has been used to quantitatively analyze the sensing mechanism by simplifying the sensor/sample system as two sets of parallel one dimensional springs arranged serially as shown schematically in Fig. 2. In previous literature, Peng et al. [21] have used a simple spring model to analyze the deflections of the two springs as a measure of the stiffness of the test tissue while the sensor transduction mechanism is based on capacitive sensing. Based on a similar concept of applying two springs with considerably different stiffnesses to soft tissue for compliance detection, Fath El-Bab et al. [22] have developed a detailed design procedure with simulations that take into account the measurement ranges of soft tissue for optimizing the sensor design parameters to give high sensitivity and linearity of the sensor output while also considering the effect of crosstalk between the two springs due to tissue deformation. They have analyzed the soft tissue stiffness as a function of the different forces experienced by the two springs with varying stiffness and have designed a micro machined piezo resistive tactile sensor [23]. In our design, we have used a two-spring model to analyze the stress distribution on the piezoelectric film from the inner and outer sensor components. Since our sensor is relatively small (Ø = 1.4), we assumed that the sensor surface can keep flat as contacted with tissue and the difference between tissue deflection when contacted by the hard and low stiffness spring is negligible to avoid crosstalk due to tissue deformation. The two springs, E1 and E2 , with differing stiffness (Young’s modulus values) represent the hard inner and soft outer components of the tactile sensor, respectively. The two identical springs Esub represent the submucosal layer while Eo refers to the object under test. The thickness of the sensor, the submucosa and the test object are denoted as Hs , Hsub and Ho , respectively. The sensor indents a uniform vertical force on the test object, resulting in localized normal stresses (1 and 2 ) which are proportional to the voltage output corresponding to its inner and outer components, respectively. In an ideal case, it is assumed that the sensor and object in contact are perfectly parallel to each other and the total vertical deformation, ␦, of the two sets of parallel springs are identical: ı = 1 + 3 + 5 = 2 + 4 + 6
(1)
where 1 and 2 correspond to the deformation of the inner (E1 ) and outer (E2 ) components of the sensor, respectively. 3 and 4
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
3
Fig. 1. Sensing mechanism of tactile sensor. (a) In contact with a soft object, the output voltage ratio is relatively small. (b) In contact with a hard object, the output voltage ratio is relatively large.
Fig. 2. The ideal contact analysis based on the tandem spring model. (a) Contact with a hard object and (b) contact with a soft object.
correspond to the deformation of the submucosa (Esub ) while 5 and 6 correspond to the deformation of the test object (Eo ). The strain experienced by the two sets of springs can be expressed by Eqs. (2)–(7):
ε2 =
2 Hs
(5)
ε4 =
4 Hsub
(6)
6 Ho
(7)
ε1 =
1 Hs
(2)
ε6 =
ε3 =
3 Hsub
(3)
In an ideal case, the normal stress can be expressed in terms of the strain (linear elastic deformation) experienced by each spring and can be given by Eqs. (8)–(13):
ε5 =
5 Ho
(4)
1 = E1
1 Hs
(8)
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS
4
C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
Fig. 4. The 2D model of a tactile sensor for numerical simulation. Table 1 The specifications of tactile sensor.
Fig. 3. The theoretical response curve of the sensor output as a function of Young’s modulus of the test object in the ideal case when the sensor and object are perfectly parallel in the presence of a normal applied force.
3 = Esub
3 Hsub
(9)
5 Ho
(10)
2 2 = E2 Hs
(11)
5 = Eo
4 = Esub 6 = Eo
4 Hsub
(12)
6 Ho
(13)
where E is the Young’s modulus and can be expressed as the ratio of stress to strain. In general, when two or more springs are arranged in series, the external stress applied to the ensemble gets applied to each spring without a change in magnitude. Thus, the stress experienced by each spring arranged serially can be equated as shown below: 1 = 3 = 5 ⇒ E1
1 3 5 = Esub = Eo Hs Ho Hsub
(14)
2 = 4 = 6 ⇒ E2
2 4 6 = Esub = Eo Hs Ho Hsub
(15)
Using Eqs. (14) and (15), 3 and 5 can be expressed in terms of 1 and 4 and 6 can be expressed in terms of 2 , respectively. By equating the total deformation of two sets of springs (Eq. 1), the ratio 1 /2 can be expressed as: 1 (EoEsubHs) + (E2EoHsub) + (E2EsubHo) = 2 (EoEsubHs) + (E1EoHsub) + (E1EsubHo)
(16)
Materials
Length (mm)
Thickness (mm)
Inner Structure (Copper Ball) Package Layer PVDF Film FPC Film Mount Elastomer Adhesive
0.4 1.4 1.4 1.4 1.4 10 0.2
0.4 0.6 28 × 10−3 50 × 10−3 1.5 10 0.2
The sensor response, expressed as the ratio of the normalized stress at the inner and outer components (1 /2 ), can then be expressed as: E1[(EsubEoHs) + (E2EoHsub) + (E2EsubHo)] 1 E1 1 = = E2 2 2 E2[(EsubEoHs) + (E1EoHsub) + (E1EsubHo)]
(17)
Thus the stress distribution on the piezoelectric film depends on the stiffness of the two sensor components and the deformation they experience when contacting tissues of varying stiffness and thickness. Simulation results of Eq. (17) (Fig. 3) show that the ratio of output voltages obtained at the inner and outer components increases as the Young’s modulus of test object increases. The sensor output shows a two-part linear dependence on the Young’s modulus of the test object below and above 1.4 MPa and this value corresponds to the Young’s modulus of the PDMS packaging. The Young’s modulus of the PDMS (E2 = 1.4 MPa) is comparable to that of soft tissue (<1 MPa) while the Young’s modulus of the copper ball (E1 = 120 GPa) is significantly higher and the ratio E1 :E2 can be estimated to be about 86,000. When contacting stiffer objects (Young’s modulus > 1.4 MPa), the majority of the compressive force is transferred by the hard copper ball and the deformation of the PDMS is small. On the contrary, the PDMS experiences a larger deformation when the sensor is contacted with objects having a Young’s modulus lower than 1.4 MPa. Thus, the determining property for the two-part linear dependence is the Young’s modulus of
Table 2 Material properties for numerical simulation.
Density (kg/m 3 ) Young’ s modulus (MPa) Poisson’s ratio Dielectric constant (Farad/m) d211 ,d233 (m/Volt) d222 (m/Volt)
PVDF
Mount
PI Film
Inner Structure
Package
Elastomer
1780 3000 0.35 11 × 10−10 −23 × 10−12 33 × 10−12
2020 23500 0.25
1353 2510 0.34
7940 110000 0.35
1030 1.32 0.4
1800 0.125–5000 0.4
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
5
Fig. 5. Numerical results of induced electric potential for an applied force of 1 N. (a) Electric potential contour under the inner structure. (b) Electric potential profile on the piezoelectric film surface corresponding to the two sensor components.
the compliant sensor component relative to that of the test object. The sensor response or sensitivity as observed by the slopes of this two-part linear dependence is higher in the lower Young’s modulus range (0.2–1.4 MPa) as compared to the higher Young’s modulus range (1.5–3.5 MPa), making it feasible for characterization of soft biological tissues. For theoretical analysis, it was observed that increasing the stiffness ratio (E1 :E2 ) of the two sensor components from 1.5:1 to 200:1 results in an increase in sensor sensitivity. This is because higher the stiffness ratio of the two sensor components, larger is the contrast in the normalized stress distribution on the PVDF film and this results in a larger difference between the voltage output obtained from the two structural electrodes. Increasing E1 :E2 to a very high value (e.g. 10,000,000), should in theory,
result in a very high sensitivity. However, besides the material limitations of achieving a very high stiffness ratio, this approach is not feasible for fabrication of practical tactile sensors. In reality, the stiffness of the test tissue should be comparable to that of the compliant sensor component and the comparison of the output voltages obtained at the two structural electrodes is only viable when both the sensor output voltages (V1 , V2 ) are detectable as a normal force is applied between the sensor and the test object. 2.3. 2D model and parameter settings A 2D model of the tactile sensor was established and calculated based on commercial finite element software, ABAQUS, as
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS
6
C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
Table 3 Different elastomers for numerical simulation.
Mucosa Dow Corning 3120 RTV Rubber Sylgard 160 Silicone Elastomer Sylgard 184 Silicone Elastomer Sylgard 170 Silicone Elastomer Sylgard 186 Silicone Elastomer
Density (kg/m3 )
Young’ s modulus (MPa)
Poisson’s ratio
980 1710 1570 1030 1380 980
0.26 3.51 3.08 1.4 1.18 0.26
0.5 0.48 0.48 0.4 0.48 0.5
which the peak value of electric potential within inner structure and the value at a distance of 1.05 mm is set as V1 and V2 , respectively. Hence, the voltage output ratios (V1 /V2 ) for five elastomers with different Young’s moduli ranging from 0.2 to 3.5 MPa can be calculated, as shown in Fig. 6. The voltage output ratio shows a two-part linear increase as a function of the Young’s modulus, thus verifying the original concept described in Fig. 1. Additionally, the tactile sensor is more sensitive in the lower Young’s modulus region (0.2–1.4 MPa) and is thus feasible for characterizing soft tissue. 2.5. Sensor fabrication
Fig. 6. Simulation results of a voltage ratio for five elastomers with different Young’s Moduli.
shown in Fig. 4. Table 1 indicates that all dimensions are the same with a finished sensor, while Table 2 lists the material properties of each part in the simulation model. A uniform external force of 1 N was set to load on the elastomers top surface within a 15 × 15 mm2 square area and the bottom surface was set as a fixedend boundary condition. The mesh was then established by 8-node hexahedral elements and the total element number was around 80,000, as decided by a convergence test. Based on this 2D model, five elastomers with different Young’s moduli were evaluated and the electric potential at the piezoelectric film was investigated. These elastomers are available from commercial products (Dow Corning Co., USA), and their Young’s moduli can be measured by Dynamic Mechanical Analyzer, (DMA instrument, TA Q800), as listed in Table 3. 2.4. Numerical simulation results The contour of electric potential at the piezoelectric film when a contact force of 1 N is applied between the sensor and elastomer is shown in Fig. 5(a). This figure reveals a plateau region of electrical potential corresponding to the stress concentration underneath the inner structure. Therefore, despite the uniformity of the applied loading, the resulting stress is non-uniform and concentrated mainly at the inner structure regime, owing to the higher stiffness of the copper ball as compared to that of the PDMS packaging. Fig. 5(b) plots the cross-sectional profile of electric potential for an elastomer E = 1.32 MPa. Consequently, the elastomer possessing a higher Young’s modulus was capable of inducing a higher electric potential at the inner structure. Next, consider a situation in
The steps involved in the fabrication of the flexible tactile sensor are schematically illustrated in Fig. 7. The flexible circuit board (FPC) which consists of a polyimide (PI) layer coated on either side with a conducting copper film is photolithographically etched to form the structural electrodes. A commercial 28 m-thick polyvinylidene fluoride (PVDF) film (Measurement Specialties, Inc.) was used as the sensing material. PVDF has attracted much interest as a piezoelectric and pyroelectric material because of its light weight, flexibility, low power consumption and non-toxicity [24,25]. The double sided Ag coating on the PVDF thin film was removed using acetone followed by rinsing with deionized water. The exposed PVDF thin film was then sandwiched between the two patterned FPC thin films and bound together using an anisotropic conductive film (ACF). The laminates of the FPC and PVDF were further stabilized using wires and made sturdy by using a bottom layer of PDMS 160. The top surface of the upper FPC with exposed copper electrode is the reference point where the copper ball is manually placed and fixed using an adhesive before being packaged in PDMS 184 by a molding technique. PDMS was chosen as the packaging material due to its attractive physical properties like flexibility and biocompatibility and relatively lower Young’s modulus as compared to other polymer materials. Furthermore, PDMS devices can be readily sterilized and are thus suitable for use as in vivo tactile sensors. 2.6. Dynamic measurement setup The piezoelectric tactile sensor cannot measure a static force as the induced electric charge dissipates rapidly. Therefore, a dynamic test system was built for analyzing the mechanical properties of elastomers and tissues, as illustrated schematically in Fig. 8. The signal generator (AFG3022, Tektronix Inc, USA) drove the shaker (GW-V4, Data Physics Co, USA) at 1 Hz and 2Vp-p to contact the test object. A force sensor (209C01, PCB Piezotronics Inc, USA) was installed on the shaker front to control the force between the tactile sensor and the object. The force sensor output signals were passed through a signal converter to an oscilloscope (AFG3022, Tektronix Inc., USA) through channel 1 (V0 ) for real-time monitoring. Additionally, two output signals of the tactile sensor were passed through a charge amplifier (B&K NEXUS2690A) to the oscilloscope through channels 2 and 3 (V1 and V2 ). For example, two voltage outputs can be obtained when the sensor comes in contact with
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
7
Fig. 7. Schematic of step by step fabrication process and an image of the miniaturized tactile sensor (Ø = 1.5 mm) mounted on an endoscope.
a sample of PDMS 160 as shown in Fig. 9. Thus, the voltage output ratio for each tested object can be calculated and information regarding its stiffness can be extracted. To create a dynamic testing system during actual MIS, the miniaturized tactile sensor is mounted at the front end of a spring wire based metal catheter that can be inserted into the additional channel present in a regular endoscope that is normally used for insertion of medical instruments for manipulation purposes or biopsies. The back end of the catheter is connected to a device that works in a similar way to a pen push button mechanism and can create an oscillatory motion for applying a dynamic force on the sensor as it contacts the test tissue as shown in Fig. 10.
Each time the button is pushed, the device exerts a normal force of about 12 N. However, due to the frictional losses and damping effects of the spring-based catheter, the actual contact force between the sensor and the tissue is measured to be 0.6 N. The two structural electrodes of the tactile sensor are connected through the metal catheter to an output port inbuilt into the device and thus the voltage output corresponding to the two sensor components could be obtained. While the endoscope can be used to look inside the organ, the tactile sensor can be used to detect tissue elasticity and thus identify the presence of tumors embedded in soft tissues.
Fig. 8. Dynamic measurement setup for analyzing output signal from tactile sensor.
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS
8
C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
Fig. 9. The voltage outputs obtained from inner (V1) and outer electrode (V2) while measuring the elasticity of silicone PDMS160.
Fig. 10. (a) The miniaturized tactile sensor (Ø = 1.4 mm) mounted at the front end of a metal catheter and electrically wired through the catheter to the back mount for obtaining the voltage output. (b) The green button is used to create a dynamic force on the sensor as it contacts the tissue during regular endoscopy. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
3. Results and discussion 3.1. Embedded elastomers mimicking submucosal tumors In order to mimic the condition of submucosal tumors, we set up a loading test bed and embedded different elastomers in the submucosa of a pig’s stomach which was obtained from the local market as shown in Fig. 11. It can be seen that the voltage out-
put ratio is directly proportional to the Young’s modulus of the elastomer (obtained by conventional tensile testing as shown in Table 4) and shows a two-part linear dependence above and below 1.4 MPa. The tactile sensor has been tested in a dynamic range of 1.01–3.51 MPa with higher sensitivity observed for elastomers with a lower Young’s modulus (1.01–1.4 MPa) as compared to elastomers having a higher Young’s modulus (1.5–3.51 MPa), further confirming the theoretical prediction made using the tandem
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
9
Fig. 11. Embedding different elastomeric materials in the submucosa of pig stomach.
Table 4 The voltage output ratio for five kinds of elastomers with varying Young’s modulus values. Elastomeric Materials
Young’s modulus (MPa)
Voltage ratio (V1 /V2 )
Standard deviation
3120 RTV Rubber Sylgard 160 Artificial Tumor Sylgard 184 Sylgard 170 Sylgard 186 Normal Tissue
3.51 3.08
2.14 2.03 1.67 1.54 1.33 1.24 1.05
±0.04 ±0.04 ±0.04 ±0.04 ±0.04 ±0.03 ±0.04
1.40 1.18 1.01
spring model and the numerical simulation results. Consequently, it is clearly demonstrated by our results that this tactile sensor was successful in differentiating the Young’s modulus of elastomers embedded in pig stomach submucosa. This further validates that elasticity differences in submucosal layers can be efficiently tested using this sensor.
3.2. Experiments for artificial submucosal tumor The efficacy of the tactile sensor can be further tested for detection of an artificial tumor that has been prepared using a common clinical method. Instant glue is injected in the underside of the pig stomach submucosa which hardens after 30 min at room temperature. The method for preparation of the artificial tumor has been shown in Fig. 13 where the normal tissue (dotted green rectangle) can be seen alongside the injected artificial tumor (dotted red circle). The tactile sensor was brought in contact with the normal tissue and the artificial tumor. The average after 10 measurements of the artificial tumor show a voltage ratio of about 1.67 which deviates significantly from the voltage ratio of 1.05 observed for normal tissue as shown in Fig. 14. Five kinds of elastomeric materials (Dow Corning Co., USA) with differing Young’s moduli were used for evaluation of the tactile sensor. Each elastomer was prepared according to the suggested mixing ratio of main and curing agent and solidified in a cubic mold (15 × 15 × 10 mm3 ). The mechanical property of the elastomer can be differentiated by the ratio of induced voltages from the piezoelectric transducers located beneath the hard copper ball (V1 ) and soft PDMS packaging (V2 ). The average voltage output ratio (V1 /V2 ) obtained for each elastomeric object after 10 measurements are shown in Fig. 12.
Fig. 12. Voltage output ratio for five kinds of elastomers with varying Young’s moduli embedded in the submucosa of pig stomach.
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS
10
C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
Fig. 13. Artificial submucosal tumor embedded in the pig stomach submucosa using a glue hardening process. (For interpretation of the references to colour in the text, the reader is referred to the web version of this article.)
Acknowledgments The authors would like to thank the Southern Taiwan Medical Device Industry Cluster, Taiwan, for financially supporting this research under Contract No. CY-13-04-30-103. The Roll-to-Roll Imprinting Center for Flexible Optoelectronics at Southern Taiwan University of Science and Technology is appreciated for the use of its MEMS fabrication facilities.
References
Fig. 14. The voltage output ratio of the tactile sensor used for differentiating artificial submucosal tumor from normal surrounding tissue in pig stomach submucosa.
4. Conclusions In summary, we have fabricated a simple piezoelectric tactile sensor for real time detection of submucosal tumors during MIS as this is often the first and most important step in cancer diagnosis and therapy. The sensing mechanism is based on the tandem spring model and the differential voltage output corresponding to the hard inner and soft outer components can provide information about tissue elasticity. Experimental results for elasticity detection of embedded elastomers and artificial tumor in pig stomach submucosa show good agreement with theoretical predictions and numerical simulations. Furthermore, this miniaturized tactile sensor can be used in conjugation with an endoscope to measure elasticity of submucosal lesions and can provide the surgeon with faster and more precise information.
[1] P. Dario, Tactile sensing: technology and applications, Sens. Actuators 1 (1) (1991) 251–256. [2] K. Hoyt, B. Castaneda, M. Zhang, P. Nigwekar, P.A. Di’sant’agnese, J.V. Joseph, J. Strang, D.J. Rubens, K.J. Parker, Tissue elasticity properties as biomarkers for prostate cancer, Cancer Biomark. 4 (4–5) (2008) 213–225. [3] T.A. Krouskop, T.M. Wheeler, F. Kallel, B.S. Garra, T. Hall, Elastic moduli of breast and prostate tissues under compression, Ultrason. Imaging 20 (4) (1998) 260–274. [4] S. Phipps, T.H.J. Yang, F.K. Habib, R.L. Reuben, S.A. McNeill, Measurement of tissue mechanical characteristics to distinguish between benign and malignant prostatic disease, Urology 66 (2) (2005) 447–450. [5] T. Wiech, A. Walch, M. Werner, Histopathological classification of nonneoplastic and neoplastic gastrointestinal submucosal lesions, Endoscopy 37 (2005) 630–634. [6] A. Chak, EUS in submucosal tumors, Gastrointest. Endosc. 56 (2002) S43–8. [7] E. Buscarini, M.D. Stasi, S. Rossi, M. Silva, F. Giangregorio, F. Giangregorio, L. Buscarini, Endosonographic diagnosis of submucosal upper gastrointestinal tract lesions and large fold gastropathies by catheter ultrasound probe, Endoscopy 29 (1997) E26. [8] D. Oguz, L. Filik, E. Parlak, S. Disibeyaz, B. Cicek, S. Kacar, G. Aydog, B. Sahin, Accuracy of endoscopic ultrasongraphy in upper gastrointestinal submucosal lesions, Turk. J. Gastroenterol. (2004) 82–85. [9] M.H. Lee, H.R. Nicholls, Tactile sensing for mechatronics—a state of the art survey, Mechatronics 9 (1999) 1–31. [10] A.T. Golpaygani, S. Najarian, M. Movahedi, Tactile sensor for robotic applications, World Congress Med. Phys. Biomed. Eng. 25 (no. 4) (2010) 2299–2302, Munich, Germany. [11] S. Omata, Y. Terunuma, New tactile sensor like the human hand and its applications, Sens. Actuators A 35 (no. 1) (1992) 9–15. [12] A. Wisitsoraat, V. Patthanasetakul, T. Lomas, A. Tuantranont, Low cost thin film based piezoresistive MEMS tactile sensor, Sens. Actuators A 139 (no. 1–2) (2007) 17–22. [13] M. Shikida, T. Shimizu, K. Sato, K. Itoigawa, Active tactile sensor for detecting contact force and hardness of an object, Sens. Actuators A 103 (no. 1–2) (2003) 213–218. [14] S. Sokhanvar, M. Packirisamy, J. Dargahi, MEMS endoscopic tactile sensor: toward in-situ and in-vivo tissue softness characterization, IEEE Sens. J. 9 (12) (2009) 1679–1687. [15] Y. Tanaka, K. Doumoto, A. Sano, H. Fujimoto, Development of a sensor system with syringe based on tactile sensing using balloon expansion, in: IEEE International Conference on Robotics and Automation, Alaska, 2010, pp. 4861–4866.
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
[16] K. Sangpradit, H. Liu, P. Dasgupta, K. Althoefer, Finite-element modeling of soft tissue rolling indentation, IEEE Trans. Biomed. Eng. (2011) 3319– 3327. [17] S. Zhao, D. Parks, C. Liu, Design and modeling of a wide dynamic-range hardness sensor for biological tissue assessment, IEEE Sens. J. (2013) 4613–4620. [18] C.H. Chuang, K.F. Lu, J.L. Mu, Flexible tactile sensor attached to laproscope for mechanical characteristics of soft materials and tissues, Transducers (2013) 2169–2172. [19] A.P. Åstrand, V. Jalkanen, B.M. Andersson, O.A. Lindahl, Detection of stiff nodules embedded in soft tissue phantoms, mimicking cancer tumours, using f tactile resonance sensor, Biomed. Sci. Eng. (2014) 181–193. [20] J. Kwon, J. Hwang, J. An, G. Yang, D. Hong, Enhanced tactile sensor for the minimally invasive robotic palpation, IEEE International Conference on Advanced Intelligent Mechatronics (2014) 1375–1380. [21] P. Peng, R. Rajamani, A.G. Erdman, Flexible tactile sensor for tissue elasticity measurements, J. Microelectromech. Syst. 18 (no. 6) (2009) 1226–1233. [22] A.M.R. Fath El Bab, T. Tamura, K. Sugano, T. Tsuchiya, O. Tabata, M.E.H. Eltaib, M.M. Sallam, Design and simulation of a tactile sensor for soft-tissue compliance detection, IEEJ Trans. Sens. Micromach. 128-E (no. 5) (2008) 186–192. [23] Ahmed M.R. Fath El Bab, Koji Sugano, Toshiyuki Tsuchiya, Osamu Tabata, Mohamed E.H. Eltaib, Mohamed M. Sallam, Micromachined tactile sensor for soft-tissue compliance detection, J. Microelectromech. Syst. 21 (3) (2012) 635–645.
11
[24] S. Cha, et al., Porous PVDF as effective sonic wave driven nanogenerators, Nano Lett. 11 (2011) 5142–5147. [25] W.J. Hu, et al., Universal ferroelectric switching dynamics of vinylidene fluoride-trifluoroethylene copolymer films, Sci. Rep. 4 (4772) (2014).
Biography
Cheng-Hsin Chuang (M’04) received his B.S. degree and Ph.D. degree from the National Cheng Kung University (NCKU) in 1995 and 2002, respectively, both in Civil Engineering. He then held the Postdoctoral research scholarship with the Center for Micro/Nano Science and Technology at NCKU, where he was in charge of the core facilities for MEMS fabrication and Nanotechnology. In 2004, he joined the Electronics Research Organization and Service (ERSO) at ITRI, where he conducted development of MEMS microphone and SAW based biosensor. Since 2005, he was recruited by Department of Mechanical Engineering and Institute of Nanotechnology at Southern Taiwan University of Science and Technology as an Assistant Professor. Currently, he is a full Professor and serves as the Director of Roll-to-Roll Imprinting Center for Flexible OptoElectronics (RicFoe) and the Micro and Nano Sensing Technology Lab (MANST Lab).
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
ARTICLE IN PRESS
SNA-9604; No. of Pages 11
Sensors and Actuators A xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Sensors and Actuators A: Physical journal homepage: www.elsevier.com/locate/sna
Piezoelectric tactile sensor for submucosal tumor detection in endoscopy Cheng-Hsin Chuang a,∗ , Tsan-Hsiu Li a , I-Chinms Chou b , Ying-Juimr Teng b a b
Department of Mechanical Engineering, Southern Taiwan University of Science and Technology, Tainan 71005, Taiwan Kuang Tai Metal Industrial Co. Ltd., Kaohsiung, Taiwan
a r t i c l e
i n f o
Article history: Received 30 September 2015 Received in revised form 6 April 2016 Accepted 6 April 2016 Available online xxx Keywords: Tactile sensor Piezoelectric Endoscopy Stiffness
a b s t r a c t In this study, we have fabricated a miniaturized piezoelectric tactile sensor (Ø = 1.4 mm) that is suitable for mounting on an endoscope to detect submucosal tumors. The sensing mechanism is based on a tandem spring model and consists of two components with varying stiffness, namely a hard inner structure embedded in a soft outer packaging. The voltage output of the PVDF sensing film is proportional to the localized normal stress exerted by each component and this differential output can be used to extract information about the elasticity of the test object/biological tissue. The sensor design has been tested by embedding different elastomers and artificial tumors in a pig’s stomach for simulating the conditions of submucosal tumors in humans and shows good agreement with theoretical analysis and numerical simulations. The sensor response is proportional to the Young’s modulus of test sample over a range of 1.01–3.51 MPa, making it suitable for detection of tumors present in soft tissues. The proposed miniaturized tactile sensor utilizes a passive sensing element, is wired to the external readout system through the metal catheter which can be inserted into the endoscopic channel, is safe for insertion into the stomach as it has a biocompatible packaging and can thus be utilized to provide tactile information during regular endoscopy. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Cancer typically manifests as hard lumps (tumors) in the soft tissues of the stomach, breast, prostate, lungs and other body organs [1]. Many studies have also demonstrated that tumors are significantly stiffer than the surrounding tissue [2–4]. Submucosal tumors (SMT) in clinical terminology are protuberant lesions or bumps covered with intact mucosa and their etiology varies from non neo-plastic lesions to true neoplasia [5,6]. SMT of the stomach are difficult to diagnose at an early stage using luminal endoscopy or barium radiography as these lesions originate in the muscularis mucosa or submucosa instead of the surface of gastric mucosa and visual detection may not be sufficient. Usually, endoscopic ultrasonography (EUS) can easily show whether a submucosal lesion is a tumor and is considered the most accurate imaging technique for the diagnosis of gastrointestinal SMT [7,8]. However, the facility of EUS is not usually available in low resource settings due to
∗ Corresponding author. E-mail address: [email protected] (C.-H. Chuang).
its high set up cost. We propose an alternative way to detect SMT based on direct contact with a miniaturized tactile sensor that is compatible with traditional endoscopy for the detection of suspicious lesions in the upper gastrointestinal tract. Tactile sensors are devices that can measure certain properties of an object or contact event through physical contact between the sensor and the object [9] and can be categorized based on the transduction mechanism that they utilize such as capacitive sensing [10], piezoelectricity [11], piezo resistivity [12] and pneumatic-based methods [13] to name a few. During MIS, much of the tactile information that is normally available during open surgery is lost. Real time measurement of elasticity or stiffness of biological tissues would help the surgeon to locate blood vessels, determine the health and type of a tissue or identify cancerous tumors whose mechanical properties are known to differ substantially from healthy tissue. Recently, many researchers have presented different medical tactile sensors with a variety of designs and principles to measure the tactile information during MIS. Sokhanva et al. [14] proposed a piezoelectric tactile sensor (size: 22 × 4 × 0.8 mm) for tissue characterization by attaching three PVDF films to two stiff supporting beams and a flexible beam to acquire the voltage output as a mea-
http://dx.doi.org/10.1016/j.sna.2016.04.020 0924-4247/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS
2
C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
sure of compliance of the object in contact. Tanaka et al. [15] proposed a real time tactile sensor system (size: Ø = 4 mm) based on balloon expansion by injection of biocompatible water and can ensure sterilization of the fluid with safe contact while requiring no electrical power. The sensor output results can effectively study both the stiffness and surface condition with potential for intraoperative brain tumor diagnosis. Sangpradit et al. [16] proposed a force feedback sensor (size: Ø = 11.5 mm) for use in MIS. Its distinguishing characteristic as compared to other tactile and force sensors is the air cushion on which the main sensing element is located. Zhao et al. [17] used the two-spring model to fabricate a tactile sensor (size: 19 × 19 mm) utilizing two piezo resistive cantilevers of varying stiffness to detect the hardness of biological tissues. Our previous work employed a tactile sensor (size: 6.5 × 3.5 mm) that can be attached to a laparoscope for differentiating soft tissues in animals [18]. The mechanical characteristics of the contact object can be differentiated by extracting the voltage outputs from the piezoelectric film at the inner and outer electrodes. Åstrand et al. [19] developed a resonance based piezoelectric tactile sensor (size: 15 mm × 5 mm) that can detect stiffness variations located up to 4 mm in silicone and chicken muscle, mimicking the presence of tumors embedded in prostate tissue. Using this tactile sensor, the shift of the resonance frequency and the force at contact with tissue can be measured and combined into a tissue stiffness parameter. Kwon et al. [20] proposed a simple tactile sensor (size: 26 mm × 8 mm) that has multiple sensing points using a semiconductor micro strain gage for recognition of tissue stiffness. The sensor exhibits high linearity with comparable spatial resolution of a human finger and can be attached to the gripper of surgical instruments for in-vivo palpation during MIS. While the aforementioned concepts can aid in improving tactile sensing during MIS, they are limited by their size and design for practical application in conjugation with regular endoscopy for obtaining information about tissue elasticity and detection of submucosal tumors (endoscopic channel Ø = 2.3 mm). An endoscope is a useful MIS tool as it not only provides an image for visual inspection but also enables biopsies and the retrieval of foreign objects from inner organs of the body that are difficult to access. However, a visual inspection for an early diagnosis of a tumor is still often insufficient and thus obtaining tactile feedback using an endoscope would be helpful for increasing the precision of examinations and thus improving the quality of treatments. In this study, we have developed a miniaturized tactile sensor that can be mounted on an endoscope and can enable the surgeon to locate and differentiate the SMT from normal tissues in the stomach. We have tested the efficacy of the proposed tactile sensor for elasticity detection of commercial elastomers and artificial tumors embedded in the submucosa of a pig’s stomach. Furthermore, the theoretical analysis in ideal contact conditions based on the tandem spring model shows good agreement with observed simulation and experimental results and can operate in a dynamic measurement range suitable for identifying tumors in soft biological tissues.
2. Sensor design 2.1. Sensing mechanism The miniaturized tactile sensor consists of two components of varying stiffness, namely a hard inner component (E1 ) consisting of a copper ball and a soft outer packaging (E2 ) made of PDMS. When the sensor contacts a test object under a uniform force, the two components undergo varying deformation depending on their stiffness in comparison to the stiffness of the test object. Since the Young’s modulus of the inner structure is greater than that of the outer packaging material (E1 > E2 ), a non-uniform stress distribu-
tion will be experienced by the piezoelectric film, resulting in two varying output voltages obtained at the corresponding structural electrodes. Since the force transfer by the two sensor components of varying stiffness is different under a uniform applied load, the charge generated (voltage output) by the piezoelectric film under the two components will also be different. When the tactile sensor contacts soft objects, the deformation of the outer packaging will be significantly greater than when the sensor contacts hard objects. Thus the force transfer by the outer structure will be relatively larger when the sensor contacts soft tissues as compared to hard objects. Consequently, there is a lower stress distribution differentiation on the piezoelectric film corresponding to the two sensor components when contacting a soft object and the ratio of the output voltages (V1 /V2 ) is relatively small as shown in Fig. 1(a). However, when contacting hard objects, the majority of the normal force will be transferred by the inner structure, resulting in a greater stress distribution differentiation and so the ratio of output voltages is greater, as shown in Fig. 1(b). Therefore, the ratio of output voltages from the inner structure and outer packaging material can be used to obtain information about the hardness (and corresponding stiffness or elasticity) of the test object. 2.2. Ideal contact analysis using the tandem spring model The tandem spring model has been used to quantitatively analyze the sensing mechanism by simplifying the sensor/sample system as two sets of parallel one dimensional springs arranged serially as shown schematically in Fig. 2. In previous literature, Peng et al. [21] have used a simple spring model to analyze the deflections of the two springs as a measure of the stiffness of the test tissue while the sensor transduction mechanism is based on capacitive sensing. Based on a similar concept of applying two springs with considerably different stiffnesses to soft tissue for compliance detection, Fath El-Bab et al. [22] have developed a detailed design procedure with simulations that take into account the measurement ranges of soft tissue for optimizing the sensor design parameters to give high sensitivity and linearity of the sensor output while also considering the effect of crosstalk between the two springs due to tissue deformation. They have analyzed the soft tissue stiffness as a function of the different forces experienced by the two springs with varying stiffness and have designed a micro machined piezo resistive tactile sensor [23]. In our design, we have used a two-spring model to analyze the stress distribution on the piezoelectric film from the inner and outer sensor components. Since our sensor is relatively small (Ø = 1.4), we assumed that the sensor surface can keep flat as contacted with tissue and the difference between tissue deflection when contacted by the hard and low stiffness spring is negligible to avoid crosstalk due to tissue deformation. The two springs, E1 and E2 , with differing stiffness (Young’s modulus values) represent the hard inner and soft outer components of the tactile sensor, respectively. The two identical springs Esub represent the submucosal layer while Eo refers to the object under test. The thickness of the sensor, the submucosa and the test object are denoted as Hs , Hsub and Ho , respectively. The sensor indents a uniform vertical force on the test object, resulting in localized normal stresses (1 and 2 ) which are proportional to the voltage output corresponding to its inner and outer components, respectively. In an ideal case, it is assumed that the sensor and object in contact are perfectly parallel to each other and the total vertical deformation, ␦, of the two sets of parallel springs are identical: ı = 1 + 3 + 5 = 2 + 4 + 6
(1)
where 1 and 2 correspond to the deformation of the inner (E1 ) and outer (E2 ) components of the sensor, respectively. 3 and 4
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
3
Fig. 1. Sensing mechanism of tactile sensor. (a) In contact with a soft object, the output voltage ratio is relatively small. (b) In contact with a hard object, the output voltage ratio is relatively large.
Fig. 2. The ideal contact analysis based on the tandem spring model. (a) Contact with a hard object and (b) contact with a soft object.
correspond to the deformation of the submucosa (Esub ) while 5 and 6 correspond to the deformation of the test object (Eo ). The strain experienced by the two sets of springs can be expressed by Eqs. (2)–(7):
ε2 =
2 Hs
(5)
ε4 =
4 Hsub
(6)
6 Ho
(7)
ε1 =
1 Hs
(2)
ε6 =
ε3 =
3 Hsub
(3)
In an ideal case, the normal stress can be expressed in terms of the strain (linear elastic deformation) experienced by each spring and can be given by Eqs. (8)–(13):
ε5 =
5 Ho
(4)
1 = E1
1 Hs
(8)
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS
4
C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
Fig. 4. The 2D model of a tactile sensor for numerical simulation. Table 1 The specifications of tactile sensor.
Fig. 3. The theoretical response curve of the sensor output as a function of Young’s modulus of the test object in the ideal case when the sensor and object are perfectly parallel in the presence of a normal applied force.
3 = Esub
3 Hsub
(9)
5 Ho
(10)
2 2 = E2 Hs
(11)
5 = Eo
4 = Esub 6 = Eo
4 Hsub
(12)
6 Ho
(13)
where E is the Young’s modulus and can be expressed as the ratio of stress to strain. In general, when two or more springs are arranged in series, the external stress applied to the ensemble gets applied to each spring without a change in magnitude. Thus, the stress experienced by each spring arranged serially can be equated as shown below: 1 = 3 = 5 ⇒ E1
1 3 5 = Esub = Eo Hs Ho Hsub
(14)
2 = 4 = 6 ⇒ E2
2 4 6 = Esub = Eo Hs Ho Hsub
(15)
Using Eqs. (14) and (15), 3 and 5 can be expressed in terms of 1 and 4 and 6 can be expressed in terms of 2 , respectively. By equating the total deformation of two sets of springs (Eq. 1), the ratio 1 /2 can be expressed as: 1 (EoEsubHs) + (E2EoHsub) + (E2EsubHo) = 2 (EoEsubHs) + (E1EoHsub) + (E1EsubHo)
(16)
Materials
Length (mm)
Thickness (mm)
Inner Structure (Copper Ball) Package Layer PVDF Film FPC Film Mount Elastomer Adhesive
0.4 1.4 1.4 1.4 1.4 10 0.2
0.4 0.6 28 × 10−3 50 × 10−3 1.5 10 0.2
The sensor response, expressed as the ratio of the normalized stress at the inner and outer components (1 /2 ), can then be expressed as: E1[(EsubEoHs) + (E2EoHsub) + (E2EsubHo)] 1 E1 1 = = E2 2 2 E2[(EsubEoHs) + (E1EoHsub) + (E1EsubHo)]
(17)
Thus the stress distribution on the piezoelectric film depends on the stiffness of the two sensor components and the deformation they experience when contacting tissues of varying stiffness and thickness. Simulation results of Eq. (17) (Fig. 3) show that the ratio of output voltages obtained at the inner and outer components increases as the Young’s modulus of test object increases. The sensor output shows a two-part linear dependence on the Young’s modulus of the test object below and above 1.4 MPa and this value corresponds to the Young’s modulus of the PDMS packaging. The Young’s modulus of the PDMS (E2 = 1.4 MPa) is comparable to that of soft tissue (<1 MPa) while the Young’s modulus of the copper ball (E1 = 120 GPa) is significantly higher and the ratio E1 :E2 can be estimated to be about 86,000. When contacting stiffer objects (Young’s modulus > 1.4 MPa), the majority of the compressive force is transferred by the hard copper ball and the deformation of the PDMS is small. On the contrary, the PDMS experiences a larger deformation when the sensor is contacted with objects having a Young’s modulus lower than 1.4 MPa. Thus, the determining property for the two-part linear dependence is the Young’s modulus of
Table 2 Material properties for numerical simulation.
Density (kg/m 3 ) Young’ s modulus (MPa) Poisson’s ratio Dielectric constant (Farad/m) d211 ,d233 (m/Volt) d222 (m/Volt)
PVDF
Mount
PI Film
Inner Structure
Package
Elastomer
1780 3000 0.35 11 × 10−10 −23 × 10−12 33 × 10−12
2020 23500 0.25
1353 2510 0.34
7940 110000 0.35
1030 1.32 0.4
1800 0.125–5000 0.4
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
5
Fig. 5. Numerical results of induced electric potential for an applied force of 1 N. (a) Electric potential contour under the inner structure. (b) Electric potential profile on the piezoelectric film surface corresponding to the two sensor components.
the compliant sensor component relative to that of the test object. The sensor response or sensitivity as observed by the slopes of this two-part linear dependence is higher in the lower Young’s modulus range (0.2–1.4 MPa) as compared to the higher Young’s modulus range (1.5–3.5 MPa), making it feasible for characterization of soft biological tissues. For theoretical analysis, it was observed that increasing the stiffness ratio (E1 :E2 ) of the two sensor components from 1.5:1 to 200:1 results in an increase in sensor sensitivity. This is because higher the stiffness ratio of the two sensor components, larger is the contrast in the normalized stress distribution on the PVDF film and this results in a larger difference between the voltage output obtained from the two structural electrodes. Increasing E1 :E2 to a very high value (e.g. 10,000,000), should in theory,
result in a very high sensitivity. However, besides the material limitations of achieving a very high stiffness ratio, this approach is not feasible for fabrication of practical tactile sensors. In reality, the stiffness of the test tissue should be comparable to that of the compliant sensor component and the comparison of the output voltages obtained at the two structural electrodes is only viable when both the sensor output voltages (V1 , V2 ) are detectable as a normal force is applied between the sensor and the test object. 2.3. 2D model and parameter settings A 2D model of the tactile sensor was established and calculated based on commercial finite element software, ABAQUS, as
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS
6
C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
Table 3 Different elastomers for numerical simulation.
Mucosa Dow Corning 3120 RTV Rubber Sylgard 160 Silicone Elastomer Sylgard 184 Silicone Elastomer Sylgard 170 Silicone Elastomer Sylgard 186 Silicone Elastomer
Density (kg/m3 )
Young’ s modulus (MPa)
Poisson’s ratio
980 1710 1570 1030 1380 980
0.26 3.51 3.08 1.4 1.18 0.26
0.5 0.48 0.48 0.4 0.48 0.5
which the peak value of electric potential within inner structure and the value at a distance of 1.05 mm is set as V1 and V2 , respectively. Hence, the voltage output ratios (V1 /V2 ) for five elastomers with different Young’s moduli ranging from 0.2 to 3.5 MPa can be calculated, as shown in Fig. 6. The voltage output ratio shows a two-part linear increase as a function of the Young’s modulus, thus verifying the original concept described in Fig. 1. Additionally, the tactile sensor is more sensitive in the lower Young’s modulus region (0.2–1.4 MPa) and is thus feasible for characterizing soft tissue. 2.5. Sensor fabrication
Fig. 6. Simulation results of a voltage ratio for five elastomers with different Young’s Moduli.
shown in Fig. 4. Table 1 indicates that all dimensions are the same with a finished sensor, while Table 2 lists the material properties of each part in the simulation model. A uniform external force of 1 N was set to load on the elastomers top surface within a 15 × 15 mm2 square area and the bottom surface was set as a fixedend boundary condition. The mesh was then established by 8-node hexahedral elements and the total element number was around 80,000, as decided by a convergence test. Based on this 2D model, five elastomers with different Young’s moduli were evaluated and the electric potential at the piezoelectric film was investigated. These elastomers are available from commercial products (Dow Corning Co., USA), and their Young’s moduli can be measured by Dynamic Mechanical Analyzer, (DMA instrument, TA Q800), as listed in Table 3. 2.4. Numerical simulation results The contour of electric potential at the piezoelectric film when a contact force of 1 N is applied between the sensor and elastomer is shown in Fig. 5(a). This figure reveals a plateau region of electrical potential corresponding to the stress concentration underneath the inner structure. Therefore, despite the uniformity of the applied loading, the resulting stress is non-uniform and concentrated mainly at the inner structure regime, owing to the higher stiffness of the copper ball as compared to that of the PDMS packaging. Fig. 5(b) plots the cross-sectional profile of electric potential for an elastomer E = 1.32 MPa. Consequently, the elastomer possessing a higher Young’s modulus was capable of inducing a higher electric potential at the inner structure. Next, consider a situation in
The steps involved in the fabrication of the flexible tactile sensor are schematically illustrated in Fig. 7. The flexible circuit board (FPC) which consists of a polyimide (PI) layer coated on either side with a conducting copper film is photolithographically etched to form the structural electrodes. A commercial 28 m-thick polyvinylidene fluoride (PVDF) film (Measurement Specialties, Inc.) was used as the sensing material. PVDF has attracted much interest as a piezoelectric and pyroelectric material because of its light weight, flexibility, low power consumption and non-toxicity [24,25]. The double sided Ag coating on the PVDF thin film was removed using acetone followed by rinsing with deionized water. The exposed PVDF thin film was then sandwiched between the two patterned FPC thin films and bound together using an anisotropic conductive film (ACF). The laminates of the FPC and PVDF were further stabilized using wires and made sturdy by using a bottom layer of PDMS 160. The top surface of the upper FPC with exposed copper electrode is the reference point where the copper ball is manually placed and fixed using an adhesive before being packaged in PDMS 184 by a molding technique. PDMS was chosen as the packaging material due to its attractive physical properties like flexibility and biocompatibility and relatively lower Young’s modulus as compared to other polymer materials. Furthermore, PDMS devices can be readily sterilized and are thus suitable for use as in vivo tactile sensors. 2.6. Dynamic measurement setup The piezoelectric tactile sensor cannot measure a static force as the induced electric charge dissipates rapidly. Therefore, a dynamic test system was built for analyzing the mechanical properties of elastomers and tissues, as illustrated schematically in Fig. 8. The signal generator (AFG3022, Tektronix Inc, USA) drove the shaker (GW-V4, Data Physics Co, USA) at 1 Hz and 2Vp-p to contact the test object. A force sensor (209C01, PCB Piezotronics Inc, USA) was installed on the shaker front to control the force between the tactile sensor and the object. The force sensor output signals were passed through a signal converter to an oscilloscope (AFG3022, Tektronix Inc., USA) through channel 1 (V0 ) for real-time monitoring. Additionally, two output signals of the tactile sensor were passed through a charge amplifier (B&K NEXUS2690A) to the oscilloscope through channels 2 and 3 (V1 and V2 ). For example, two voltage outputs can be obtained when the sensor comes in contact with
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
7
Fig. 7. Schematic of step by step fabrication process and an image of the miniaturized tactile sensor (Ø = 1.5 mm) mounted on an endoscope.
a sample of PDMS 160 as shown in Fig. 9. Thus, the voltage output ratio for each tested object can be calculated and information regarding its stiffness can be extracted. To create a dynamic testing system during actual MIS, the miniaturized tactile sensor is mounted at the front end of a spring wire based metal catheter that can be inserted into the additional channel present in a regular endoscope that is normally used for insertion of medical instruments for manipulation purposes or biopsies. The back end of the catheter is connected to a device that works in a similar way to a pen push button mechanism and can create an oscillatory motion for applying a dynamic force on the sensor as it contacts the test tissue as shown in Fig. 10.
Each time the button is pushed, the device exerts a normal force of about 12 N. However, due to the frictional losses and damping effects of the spring-based catheter, the actual contact force between the sensor and the tissue is measured to be 0.6 N. The two structural electrodes of the tactile sensor are connected through the metal catheter to an output port inbuilt into the device and thus the voltage output corresponding to the two sensor components could be obtained. While the endoscope can be used to look inside the organ, the tactile sensor can be used to detect tissue elasticity and thus identify the presence of tumors embedded in soft tissues.
Fig. 8. Dynamic measurement setup for analyzing output signal from tactile sensor.
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS
8
C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
Fig. 9. The voltage outputs obtained from inner (V1) and outer electrode (V2) while measuring the elasticity of silicone PDMS160.
Fig. 10. (a) The miniaturized tactile sensor (Ø = 1.4 mm) mounted at the front end of a metal catheter and electrically wired through the catheter to the back mount for obtaining the voltage output. (b) The green button is used to create a dynamic force on the sensor as it contacts the tissue during regular endoscopy. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
3. Results and discussion 3.1. Embedded elastomers mimicking submucosal tumors In order to mimic the condition of submucosal tumors, we set up a loading test bed and embedded different elastomers in the submucosa of a pig’s stomach which was obtained from the local market as shown in Fig. 11. It can be seen that the voltage out-
put ratio is directly proportional to the Young’s modulus of the elastomer (obtained by conventional tensile testing as shown in Table 4) and shows a two-part linear dependence above and below 1.4 MPa. The tactile sensor has been tested in a dynamic range of 1.01–3.51 MPa with higher sensitivity observed for elastomers with a lower Young’s modulus (1.01–1.4 MPa) as compared to elastomers having a higher Young’s modulus (1.5–3.51 MPa), further confirming the theoretical prediction made using the tandem
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
9
Fig. 11. Embedding different elastomeric materials in the submucosa of pig stomach.
Table 4 The voltage output ratio for five kinds of elastomers with varying Young’s modulus values. Elastomeric Materials
Young’s modulus (MPa)
Voltage ratio (V1 /V2 )
Standard deviation
3120 RTV Rubber Sylgard 160 Artificial Tumor Sylgard 184 Sylgard 170 Sylgard 186 Normal Tissue
3.51 3.08
2.14 2.03 1.67 1.54 1.33 1.24 1.05
±0.04 ±0.04 ±0.04 ±0.04 ±0.04 ±0.03 ±0.04
1.40 1.18 1.01
spring model and the numerical simulation results. Consequently, it is clearly demonstrated by our results that this tactile sensor was successful in differentiating the Young’s modulus of elastomers embedded in pig stomach submucosa. This further validates that elasticity differences in submucosal layers can be efficiently tested using this sensor.
3.2. Experiments for artificial submucosal tumor The efficacy of the tactile sensor can be further tested for detection of an artificial tumor that has been prepared using a common clinical method. Instant glue is injected in the underside of the pig stomach submucosa which hardens after 30 min at room temperature. The method for preparation of the artificial tumor has been shown in Fig. 13 where the normal tissue (dotted green rectangle) can be seen alongside the injected artificial tumor (dotted red circle). The tactile sensor was brought in contact with the normal tissue and the artificial tumor. The average after 10 measurements of the artificial tumor show a voltage ratio of about 1.67 which deviates significantly from the voltage ratio of 1.05 observed for normal tissue as shown in Fig. 14. Five kinds of elastomeric materials (Dow Corning Co., USA) with differing Young’s moduli were used for evaluation of the tactile sensor. Each elastomer was prepared according to the suggested mixing ratio of main and curing agent and solidified in a cubic mold (15 × 15 × 10 mm3 ). The mechanical property of the elastomer can be differentiated by the ratio of induced voltages from the piezoelectric transducers located beneath the hard copper ball (V1 ) and soft PDMS packaging (V2 ). The average voltage output ratio (V1 /V2 ) obtained for each elastomeric object after 10 measurements are shown in Fig. 12.
Fig. 12. Voltage output ratio for five kinds of elastomers with varying Young’s moduli embedded in the submucosa of pig stomach.
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS
10
C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
Fig. 13. Artificial submucosal tumor embedded in the pig stomach submucosa using a glue hardening process. (For interpretation of the references to colour in the text, the reader is referred to the web version of this article.)
Acknowledgments The authors would like to thank the Southern Taiwan Medical Device Industry Cluster, Taiwan, for financially supporting this research under Contract No. CY-13-04-30-103. The Roll-to-Roll Imprinting Center for Flexible Optoelectronics at Southern Taiwan University of Science and Technology is appreciated for the use of its MEMS fabrication facilities.
References
Fig. 14. The voltage output ratio of the tactile sensor used for differentiating artificial submucosal tumor from normal surrounding tissue in pig stomach submucosa.
4. Conclusions In summary, we have fabricated a simple piezoelectric tactile sensor for real time detection of submucosal tumors during MIS as this is often the first and most important step in cancer diagnosis and therapy. The sensing mechanism is based on the tandem spring model and the differential voltage output corresponding to the hard inner and soft outer components can provide information about tissue elasticity. Experimental results for elasticity detection of embedded elastomers and artificial tumor in pig stomach submucosa show good agreement with theoretical predictions and numerical simulations. Furthermore, this miniaturized tactile sensor can be used in conjugation with an endoscope to measure elasticity of submucosal lesions and can provide the surgeon with faster and more precise information.
[1] P. Dario, Tactile sensing: technology and applications, Sens. Actuators 1 (1) (1991) 251–256. [2] K. Hoyt, B. Castaneda, M. Zhang, P. Nigwekar, P.A. Di’sant’agnese, J.V. Joseph, J. Strang, D.J. Rubens, K.J. Parker, Tissue elasticity properties as biomarkers for prostate cancer, Cancer Biomark. 4 (4–5) (2008) 213–225. [3] T.A. Krouskop, T.M. Wheeler, F. Kallel, B.S. Garra, T. Hall, Elastic moduli of breast and prostate tissues under compression, Ultrason. Imaging 20 (4) (1998) 260–274. [4] S. Phipps, T.H.J. Yang, F.K. Habib, R.L. Reuben, S.A. McNeill, Measurement of tissue mechanical characteristics to distinguish between benign and malignant prostatic disease, Urology 66 (2) (2005) 447–450. [5] T. Wiech, A. Walch, M. Werner, Histopathological classification of nonneoplastic and neoplastic gastrointestinal submucosal lesions, Endoscopy 37 (2005) 630–634. [6] A. Chak, EUS in submucosal tumors, Gastrointest. Endosc. 56 (2002) S43–8. [7] E. Buscarini, M.D. Stasi, S. Rossi, M. Silva, F. Giangregorio, F. Giangregorio, L. Buscarini, Endosonographic diagnosis of submucosal upper gastrointestinal tract lesions and large fold gastropathies by catheter ultrasound probe, Endoscopy 29 (1997) E26. [8] D. Oguz, L. Filik, E. Parlak, S. Disibeyaz, B. Cicek, S. Kacar, G. Aydog, B. Sahin, Accuracy of endoscopic ultrasongraphy in upper gastrointestinal submucosal lesions, Turk. J. Gastroenterol. (2004) 82–85. [9] M.H. Lee, H.R. Nicholls, Tactile sensing for mechatronics—a state of the art survey, Mechatronics 9 (1999) 1–31. [10] A.T. Golpaygani, S. Najarian, M. Movahedi, Tactile sensor for robotic applications, World Congress Med. Phys. Biomed. Eng. 25 (no. 4) (2010) 2299–2302, Munich, Germany. [11] S. Omata, Y. Terunuma, New tactile sensor like the human hand and its applications, Sens. Actuators A 35 (no. 1) (1992) 9–15. [12] A. Wisitsoraat, V. Patthanasetakul, T. Lomas, A. Tuantranont, Low cost thin film based piezoresistive MEMS tactile sensor, Sens. Actuators A 139 (no. 1–2) (2007) 17–22. [13] M. Shikida, T. Shimizu, K. Sato, K. Itoigawa, Active tactile sensor for detecting contact force and hardness of an object, Sens. Actuators A 103 (no. 1–2) (2003) 213–218. [14] S. Sokhanvar, M. Packirisamy, J. Dargahi, MEMS endoscopic tactile sensor: toward in-situ and in-vivo tissue softness characterization, IEEE Sens. J. 9 (12) (2009) 1679–1687. [15] Y. Tanaka, K. Doumoto, A. Sano, H. Fujimoto, Development of a sensor system with syringe based on tactile sensing using balloon expansion, in: IEEE International Conference on Robotics and Automation, Alaska, 2010, pp. 4861–4866.
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020
G Model SNA-9604; No. of Pages 11
ARTICLE IN PRESS C.-H. Chuang et al. / Sensors and Actuators A xxx (2016) xxx–xxx
[16] K. Sangpradit, H. Liu, P. Dasgupta, K. Althoefer, Finite-element modeling of soft tissue rolling indentation, IEEE Trans. Biomed. Eng. (2011) 3319– 3327. [17] S. Zhao, D. Parks, C. Liu, Design and modeling of a wide dynamic-range hardness sensor for biological tissue assessment, IEEE Sens. J. (2013) 4613–4620. [18] C.H. Chuang, K.F. Lu, J.L. Mu, Flexible tactile sensor attached to laproscope for mechanical characteristics of soft materials and tissues, Transducers (2013) 2169–2172. [19] A.P. Åstrand, V. Jalkanen, B.M. Andersson, O.A. Lindahl, Detection of stiff nodules embedded in soft tissue phantoms, mimicking cancer tumours, using f tactile resonance sensor, Biomed. Sci. Eng. (2014) 181–193. [20] J. Kwon, J. Hwang, J. An, G. Yang, D. Hong, Enhanced tactile sensor for the minimally invasive robotic palpation, IEEE International Conference on Advanced Intelligent Mechatronics (2014) 1375–1380. [21] P. Peng, R. Rajamani, A.G. Erdman, Flexible tactile sensor for tissue elasticity measurements, J. Microelectromech. Syst. 18 (no. 6) (2009) 1226–1233. [22] A.M.R. Fath El Bab, T. Tamura, K. Sugano, T. Tsuchiya, O. Tabata, M.E.H. Eltaib, M.M. Sallam, Design and simulation of a tactile sensor for soft-tissue compliance detection, IEEJ Trans. Sens. Micromach. 128-E (no. 5) (2008) 186–192. [23] Ahmed M.R. Fath El Bab, Koji Sugano, Toshiyuki Tsuchiya, Osamu Tabata, Mohamed E.H. Eltaib, Mohamed M. Sallam, Micromachined tactile sensor for soft-tissue compliance detection, J. Microelectromech. Syst. 21 (3) (2012) 635–645.
11
[24] S. Cha, et al., Porous PVDF as effective sonic wave driven nanogenerators, Nano Lett. 11 (2011) 5142–5147. [25] W.J. Hu, et al., Universal ferroelectric switching dynamics of vinylidene fluoride-trifluoroethylene copolymer films, Sci. Rep. 4 (4772) (2014).
Biography
Cheng-Hsin Chuang (M’04) received his B.S. degree and Ph.D. degree from the National Cheng Kung University (NCKU) in 1995 and 2002, respectively, both in Civil Engineering. He then held the Postdoctoral research scholarship with the Center for Micro/Nano Science and Technology at NCKU, where he was in charge of the core facilities for MEMS fabrication and Nanotechnology. In 2004, he joined the Electronics Research Organization and Service (ERSO) at ITRI, where he conducted development of MEMS microphone and SAW based biosensor. Since 2005, he was recruited by Department of Mechanical Engineering and Institute of Nanotechnology at Southern Taiwan University of Science and Technology as an Assistant Professor. Currently, he is a full Professor and serves as the Director of Roll-to-Roll Imprinting Center for Flexible OptoElectronics (RicFoe) and the Micro and Nano Sensing Technology Lab (MANST Lab).
Please cite this article in press as: C.-H. Chuang, et al., Piezoelectric tactile sensor for submucosal tumor detection in endoscopy, Sens. Actuators A: Phys. (2016), http://dx.doi.org/10.1016/j.sna.2016.04.020