1.6 Thesis Organization
1.6.7 Chapter 7 — Conclusions
This chapter briefly summarizes all the work presented in this thesis and offers recom- mendations for the future work necessary to eventually take the new sensors and instru- ments described in this thesis to clinical evaluation.
References
[1] B. Stewart and C. P. Wild, Eds.,World Cancer Report 2014. International Agency
References 23
[2] A. R. Lanfranco, A. E. Castellanos, J. P. Desai, and W. C. Meyers, “Robotic surgery: A current perspective,”Annals of Surgery, vol. 239, no. 1, pp. 14–21, Jan. 2004.
[3] G. J. Tortora and B. Derrickson,Principles of anatomy and physiology, 11th ed. John
Wiley & Sons, 2006.
[4] Wikimedia Commons, “Lungs diagram detailed,” Available online: https://commons.wikimedia.org/wiki/File:Lungs_diagram_detailed.svg, Accessed: 27 May 2015.
[5] J. A. Roth, J. C. Ruckdeschel, and T. H. Weisenburger, Thoracic oncology, 2nd ed.
Philadelphia, Pennsylvania: W. B. Saunders Company, 1995.
[6] D. Mahalingam, A. Mita, M. M. Mita, S. T. Nawrocki, and F. J. Giles, “Targeted ther- apy for advanced non-small cell lung cancers: historical perspective, current practices, and future development,”Curr. Prob. Cancer, vol. 33, no. 2, p. 65–112, Mar. 2009.
[7] Canadian Cancer Society, “Lung cancer statistics,” Available online: http://www.cancer.ca/en/cancer-information/cancer-type/lung/statistics/, Accessed: 25 May 2015.
[8] American Cancer Society, “Non-small cell lung cancer survival rates by stage,” Available online: http://www.cancer.org/cancer/lungcancer-non- smallcell/detailedguide/non-small-cell-lung-cancer-survival-rates, Accessed: 25 May 2015.
[9] S. R. Desai, T. Franquet, T. E. Hartman, and A. U. Wells,Pulmonary imaging: Con-
tributions to Key Clinical Questions. Boca Raton, Florida: Informa UK Ltd., 2007.
[10] L. Reveiz, J. R. Rueda, and A. F. Cardona, “Palliative endobronchial brachytherapy for non-small cell lung cancer,”Cochrane Database Syst. Rev., Dec. 2012.
References 24
[11] B. Hilaris and D. A. Mastoras, “Contemporary brachytherapy approaches in non small-cell lung cancer,”J. Surg. Oncol., vol. 69, p. 258–264, Dec. 1998.
[12] J. P. van Meerbeeck and V. F. M. Surmont, “Stage IIIA-N2 NSCLC: A review of its treatment approaches and future developments,” Lung Cancer, vol. 65, no. 3, p.
257–267, Sep. 2009.
[13] B. Jeremic and M. Bamberg, “External beam radiation therapy for bronchial stump recurrence of non-small-cell lung cancer after complete resection,”Radiotherapy and
Oncology, vol. 64, no. 3, p. 251–257, Sep. 2002.
[14] H. Nakamura, “Controversies in thoracoscopic lobectomy for lung cancer: editorial,”
Ann. Thorac. Cardiovasc. Surg., vol. 13, no. 4, pp. 225–227, Sep. 2007.
[15] R. H. Bell and D. B. Kaufman, Northwestern handbook of surgical procedures.
Georgetown, Texas: Landes Bioscience, 2005.
[16] R. Gharagozloo and F. Najam,Robotic Surgery. China: McGraw-Hill Companies
Inc., 2009.
[17] D. Cheng, R. J. Downey, K. Kernstine, R. Stanbridge, H. Shennib, R. Wolf, T. Oht- suka, R. Schmid, D. Waller, H. Fernando, A. Yim, and J. Martin, “Video-assisted thoracic surgery in lung cancer resection: a meta-analysis and systematic review of controlled trials,”Innovations (Philadelphia), vol. 2, no. 6, pp. 261–292, Nov. 2007.
[18] T. L. Demmy and J. J. Curris, “Minimally invasive lobectomy directed toward frail and high-risk patients: A case control study,”Ann. Thorac. Surg., vol. 68, no. 1, pp.
References 25
[19] D. Sortini, C. V. Feo, P. Carcoforo, G. Carrella, E. Pozza, A. Liboni, and A. Sortini, “Thoracoscopic localization techniques for patients with solitary pulmonary nodule and history of malignancy,”Ann. Thorac. Surg., vol. 79, no. 1, p. 258–262, Jan. 2005.
[20] M. Kaneko, C. Toya, and M. Okajima, “Active strobe imager for visualizing dynamic behavior of tumours,” inProc. IEEE Int. Conf. Robot. Autom., Roma, Italy, May 2007,
p. 3009–3014.
[21] P. Ciriaco, G. Negri, A. Puglisi, R. Nicoletti, A. D. Maschio, and P. Zannini, “Video- assisted thoracoscopic surgery for pulmonary nodules: rationale for preoperative computed tomography-guided hookwire localization,”Euro. J. Cardio. Thorac. Surg.,
vol. 25, no. 3, p. 429–433, Mar. 2004.
[22] J. Hu, C. Zhang, and L. Sun, “Localization of small pulmonary nodules for videotho- racoscopic surgery,”Anz. J. Surg., vol. 76, no. 7, p. 649–651, Jul. 2006.
[23] R. Santambrogio, M. Montorsi, P. Bianchi, A. Mantovani, F. Ghelma, and M. Mezzetti, “Intraoperative ultrasound during thoracoscopic procedures for solitary pulmonary nodules,”Ann. Thorac. Surg., vol. 68, no. 1, pp. 218–222, Jul. 1999.
[24] R. H. Daffner,Clinical radiology: The essentials. Baltimore, Maryland: Williams & Wilkins, 1993.
[25] S. Matsumoto, T. Hirata, E. Ogawa, T. Fukuse, H. Ueda, T. Koyama, T. Nakamura, and H. Wada, “Ultrasonographic evaluation of small nodules in the peripheral lung during video-assisted thoracic surgery,”Euro. J. Cardio. Thorac. Surg., vol. 26, no. 3,
p. 469–473, Sep. 2004.
[26] V. D. M. Hornblower, E. Yu, A. Fenster, J. J. Battista, and R. A. Malthaner, “3D thoracoscopic ultrasound volume measurement validation in an ex vivo and in vivo
References 26
porcine model of lung tumors,” Phys. Med. Biol., vol. 52, no. 91, p. 91–106, Jan.
2007.
[27] A. P. Miller, W. J. Peine, J. S. Son, and Z. T. Hammoud, “Tactile imaging system for localizing lung nodules during video assisted thoracoscopic surgery,” inProc. IEEE
Int. Conf. Robot. Autom., Rome, Italy, Apr. 2007, pp. 2996–3001.
[28] R. H. Taylor and D. Stoianovici, “Medical robotics in computer integrated surgery,”
IEEE Trans. Robot. Autom., vol. 19, no. 5, p. 765–781, Oct. 2003.
[29] F. M. A. Melfi, G. F. Menconi, A. M. Mariani, and C. A. Angeletti, “Early experience with robotic technology for thoracoscopic surgery,”Euro. J. Cardio. Thorac. Surg.,
vol. 21, no. 5, pp. 864–868, May 2002.
[30] B. Deml, T. Ortmaier, and U. Seibold, “The touch and feel in minimally invasive surgery,” inProc. IEEE Int. Workshop Haptic Aud. Vis. Env. Appl., Oct. 2005, p. 6.
[31] S. M. Hosseini, S. Najarian, S. Motaghinasab, A. T. Golpaygani, and S. Torabi, “Pre- diction of tumor existence in the virtual soft tissue by using tactile tumor detector,”
Amer. J. Appl. Sci., vol. 5, no. 5, pp. 483–489, May 2008.
[32] M. Ottermo, M. Ovstedal, T. Lango, O. Stavdahl, Y. Yavuz, T. Johansen, and R. Mar- vik, “The role of tactile feedback in laparoscopic surgery,” Surgical Laparoscopy,
Endoscopy&Percutaneous Techniques, vol. 16, no. 6, pp. 390–400, Dec. 2006. [33] J. Konstantinova, A. Jiang, K. Althoefer, P. Dasgupta, and T. Nanayakkara, “Im-
plementation of tactile sensing for palpation in robot-assisted minimally invasive surgery: A review,”IEEE Sensors J., vol. 14, no. 8, pp. 2490–2501, Aug. 2014.
Chapter 2
Literature Review
2.1
Tactile Sensors
The use of a tactile sensor array has been explored in the past to enable virtual tactile feedback for palpation during MIS [1]. A tactile sensor array (or simply, tactile sensor) is a cluster of small discrete contact pressure sensing elements that can be used to detect varia- tions in underlying stiffness when pressed against a surface. Since tumours have a different stiffness when compared to the surrounding soft tissue, a tactile sensor can be used to locate them [2]. There are several research groups and commercial organizations that have devel- oped different tactile sensors for MIS, however they have some major drawbacks, including lack of sterilizability, using a large number of wires (preventing their use in robotic MIS instruments with a wrist), and high cost due to expensive and highly specialized manufac- turing processes [3]. Hence, developing a low-cost, re-sterilizable/disposable tactile sensor with a limited number of wires is paramount in enabling the clinical use of this technology. A thorough review of existing tactile sensor designs was conducted to determine the state- of-the-art. The three major design choices that can be used to classify tactile sensors are the sensing technology employed, the configuration used and the fabrication process.
2.1 Tactile Sensors 28
2.1.1
Sensing Technologies
Several different pressure sensing technologies have been explored by researchers to construct tactile sensors. The key requirement for a pressure sensing technology to be feasible for use in tactile sensors is the ability to make the individual sensing elements small enough to provide reasonable spatial resolution. It is generally accepted that a spatial resolution of 3×3 mm is a recommended minimum for obtaining any useful tactile infor- mation [4]. Depending on the application, the spatial resolution can be as fine as 1×1 mm in some existing tactile sensor designs. Almost all of the existing tactile sensors use one of the following sensing technologies.
Piezoelectric: This approach uses the measurement of either the charge created across
a piezoelectric material, or the change in the electrical impedance of the material to deter- mine the applied pressure [5–8]. Piezoelectric materials typically used in tactile sensors are Lead Zirconate Titanate (PZT, rigid) crystals and Polyvinylidene Fluoride (PVDF, flexible) films. PVDF is typically preferred due to its flexibility and low-cost, but PZT provides much better sensitivity. The measurement of the generated charge is simpler than the mea- surement of the change in electrical impedance. However, the latter typically provides more accurate measurement under quasi-static loads, which is the normal mode of operation for tactile sensors since they are pressed and held against the target tissue for a few seconds. The main drawback of piezoelectric sensing is the complexity of the circuitry required to make accurate measurements.
Piezoresistive: This approach uses the phenomenon that some materials experience a
decrease in their electrical resistance on the application of pressure [9–13]. Some piezo- resistive materials are fabricated by evenly dispersing fine particles of a conductive material such as silver, nickel, carbon black or graphite in a polymer or elastomer matrix [14]. The concentration of the conductive particles in the composite is adjusted to obtain a volume
2.1 Tactile Sensors 29
resistivity of a few hundredΩ-cm. The relationship between the resistance of these com- posites and the applied pressure depends on the size, material and concentration of the conductive particles, and the material of the matrix. Depending on the mechanical prop- erties of the matrix material, the measurement can have significant hysteresis and a long response time. Another piezoresistive material that is used in some MEMS piezoresistive tactile sensors is Indium Tin Oxide (ITO), a heavily-doped n-type semiconductor that is typically deposited directly on the sensor using vapour deposition or layer etching [15–18]. ITO is a stiffer material and is used in thinner layers when compared to piezoresistive poly- mers and elastomers. The change in resistance can be measured using a Wheatstone bridge or a simple voltage divider.
Capacitive: This approach uses the change in the electrical capacitance of an electro-
mechanical structure in response to applied pressure [19, 20]. The basic principle is that the capacitance between two electrodes changes as the properties or dimensions of the dielectric material between the electrodes varies as a function of applied pressure. The dielectric material can be air, polymers, elastomers or even oils [21]. The capacitance typically increases with an increase in applied pressure in most existing capacitive sensor designs. The measurement of the capacitance is usually performed by measuring the time taken to charge and/or discharge the electro-mechanical capacitor through an R-C circuit with a known resistance. Some of the notable commercial tactile sensors, such as the TactArray sensor from Pressure Profile Systems (PPS) [22] and the SureTouch sensor from Medical Tactile Inc. [23], also use capacitive sensing with excellent results, suggesting that capacitive tactile sensors can be very sensitive. The drawbacks of these commercial sensors are a large number of signal wires and high cost. These sensors are marketed for aiding in breast tumour localization by external palpation, but some researchers have adapted them to be used for MIS.
2.1 Tactile Sensors 30
applied pressure. Most designs use optical fibres to carry the light to and from the sens- ing elements so that the electronics can be placed away from the sensor [24]. The main advantage offered by optical sensors is added the safety and robustness that results from avoiding the use of electricity in the sensor. The challenges, however, are the complexity of manufacturing the optical structures and the management of the optical fibre bundles that can become bulky with a large number of sensing elements. Some optical techniques employed in tactile sensors include using fibre Bragg grating arrays for measuring pressure through strain, the measurement of bending losses in optical fibres due to applied pressure, and using structures that alter the amount of reflected light based on the applied pressure.
2.1.2
Configuration
The configuration of a tactile sensor dictates how the discrete sensing elements are linked together to form an array. The majority of tactile sensors either use independent elements or a mechanical multiplexing configuration. There are only a few designs that have employed electrical multiplexing.
Independent Elements: This is the simplest configuration in which the individual sens-
ing elements are completely independent pressure sensing units. Most optical tactile sen- sors [24], and some piezoelectric [5–8], resistive [9, 10, 14–18] and capacitive sensors [19] use this configuration. This configuration has negligible crosstalk, but results in the highest number of interface lines, since each element requires one or two optical fibres or wires. The large number of interface lines does not allow these sensors to be used on instruments with an actuated wrist.
Mechanical Multiplexing: This is a widely used configuration for electrical tactile
sensors in which the sensing elements share electrodes as a feature of the mechanical con- struction in order to greatly decrease the number of interface lines [11, 12, 20–23]. Two
2.1 Tactile Sensors 31
popular configurations are common electrode and grid electrode. In the common electrode configuration, one of the electrodes of the sensing elements is shared by all of the elements, which is typically the ground electrode. This reduces the number of interface lines for N
sensing elements to N+1 instead of 2N. In the grid electrode configuration, one of the electrodes is shared by the elements in the same row while the other is shared by elements in the same column. By selecting the correct row–column pair, each element can be indi- vidually accessed. This reduces the number of interface lines for M×N sensing elements
toM+Ninstead of 2MN. This form of multiplexing is used by the commercial TactArray and SureTouch tactile sensors.
Electrical Multiplexing: This configuration employs the use of some electrical multi-
plexing circuitry on board the tactile sensor to significantly reduce the number of interface lines. This approach decreases the number of measurement lines to only a few by adding some digital selection lines depending on the extent of multiplexing. This novel approach is employed by very few existing designs. The most notable design that uses electrical mul- tiplexing is the one by Schosteket al.[13] that uses an on board 32:1 analog multiplexer on
the rigid LTCC (low temperature co-fired ceramic) sensor to reduce the number of output lines to only five digital and one analog for 32 sensing elements. This design sacrifices the flexibility of the sensor in exchange for a compact interface. In MIS applications, the tac- tile sensor is typically mounted on a rigid tool, hence the flexibility of the sensor is not an essential feature. The design also uses distinct multiplexed sensing elements instead of the grid design, thereby reducing crosstalk and simplifying the construction. The drawback of the design is that there are still six output lines for only 32 elements, with one of the lines being analog. It is not preferable to have long analog lines due to noise considera- tions. Also, the LTCC technology is not a common manufacturing process and is relatively expensive for small production quantities.
2.1 Tactile Sensors 32
2.1.3
Fabrication
There are three common fabrication processes found in the literature for the manufac- ture of tactile sensors. The selection of the fabrication process depends on the mechanical design, and dictates the majority of the cost of the sensor. A tactile sensor may utilize one or more of the processes listed below.
MEMS: Micro-Electro-Mechanical Systems is a relatively new class of devices that
involve electro-mechanical structures made of semiconductors, polymers, metals and ce- ramics at the scale of 20µm to 1 mm. MEMS has become very popular in the manufacture of compact sensors such as accelerometers and gyroscopes that can be packaged in a single IC. Due to the compactness and versatility of MEMS, researchers have investigated its use in tactile sensing [7, 8, 13, 15–18]. MEMS devices use similar process technology as semi- conductor devices, such as deposition of material layers, patterning by photo-lithography, chemical etching, co-fired ceramics, etc. The challenge of developing MEMS tactile sen- sors is that it requires access to very specialized equipment and significant custom process design, which makes the sensors very expensive for small scale production. However, MEMS has many benefits in large scale production such has high precision and repeatabil- ity. MEMS tactile sensors found in the literature are usually rigid.
Layer Deposition: Layer deposition is a chemical manufacturing process that can be
employed to manufacture tactile sensors that use a layered design involving polymers and metals [6, 14, 19–21]. It is much simpler than MEMS but still requires the use of special- ized equipment and custom developed processes. The layers are typically deposited using foil adhesion and etching for metals, and vapour deposition for polymers. This process is commonly used to manufacture flexible tactile sensors.
Macro-scale: Macro-scale manufacturing is the simplest of all manufacturing pro-
2.2 Miniature Force/Torque Sensors 33