Minimally invasive surgery is often preferred to an open procedure to reduce trauma, haemorrhage and recovery time for the patient. The diagnosis and localization of liver tumors, however, is performed by stroking across the surface of the diseased organ as the cancerous nodes can be felt as stiff inclusions within the soft, healthy tissue. Due to the sealed peritoneum with only small access ports at the trocars this method cannot be adopted in laparoscopy. Therefore, tactile probes have been developed by different research groups, e.g. Trejos  and Schostek . However, they all exploit merely the limited space determined by the diameter of a 10mm trocar as the sensor unit of these devices is simply placed at the tip of a long slender probe without a mechanism for enlargement of the palpation area.
Therefore, a laparoscopic instrument equipped with a laterally opening lid has been developed to provide a large sensitive area for palpation of hepatic cancerous lumps. Mechanical requirements have been determined as guidelines for the design. Thereby, the most important target was to create a large sensing area despite the size constraints of a 10mm trocar. Furthermore, it was prescribed to keep the mechanics as simple as possible to maintain low effort and costs for manufacturing. Storage for the sensor unit plus the associated electronics had to be provided within the concept and, additionally, the design had to combine compliance and rigidity, thus, a solid structure around the flexible sensor as palpation would otherwise not be possible.
The prototypical version laser sintered in polyamide (PA) is pictured in Figure 1. Its diameter is 10 mm in the closed state so that it fits through a standard trocar. However, when opened it provides a tactile surface of 18 mm × 20 mm which is unique in terms of minimally invasive tactile tools. The prototype could already be used for test series to evaluate the suitability of the innovative construction.
During the experimental phase automated and manual subject palpation tests were performed to evaluate the innovative instrument design which was selected out of various concepts. Therefore, a flexible silicone sensor developed by Strohmayr  was mounted to the tip of the prototypical tool. The sensory system consists of two silicone foil substrates which both carry parallel polymer based circuit tracks (PBCTs) including conductive particles. By merging the two layers the orthogonally arranged PBCTs create crossing points, so called tactile sensing elements or taxels. Thus, an externally applied force can be measured according to the decrease of resistivity between the approximated conductive lines as it can be seen in Figure 2. For restoring the original state of the sensor after an indentation polymer spacers are located within the empty spaces of the PBCT lattice structure.
For the testing phase the sensor was attached to the mechanical part of the instrument by use of a nonconductive adhesive tape in order to avoid unintended contact of the PBCTs among each other. For the automated test series the prototype was mounted onto a linear rail. Tumor samples were created on the base of the research of Fröhlich  by equipping soft silicone as well as porcine liver specimen with hard silicone nodules. They were placed directly underneath the tactile area of the tool and on top of a scale so that the forces which have been applied during the palpation tests could subsequently be calculated from the gathered weight data. The instrument was automatically driven down onto the tumor samples for palpation as it can be seen in Figure 3. Thereby, the sensor signals were recorded so that they could later be processed in MATLAB and displayed within a 3D bar graph array for better visualization of the effects occurring at the sensory tool tip. Finally, the data has been evaluated and compared for different tumor sample sizes and positions.
The setup for the manually performed palpation tests is pictured in Figure 4. Subjects were asked to explore liver samples equipped with three or four artificial tumors using the tactile instrument. Their task was to determine the number of hidden lumps by interpreting the sensor signal displayed on a screen in front of them in real time. Moreover, they were asked to rate their confidence about the detection of a tumor and to assess the handling of the tool. In addition, the time which elapsed until the detection of a lump was noted for each candidate. The samples have again been placed on top of a scale so that the forces applied during manual palpation could be calculated from the collected weight data after the tests were completed.
The sensor data recorded during the automated experiments was evaluated and displayed as a 3D color-coded bar graph. Blue color stands for low, red for high external load, thus, forces applied onto the sensor through palpation. As higher forces occur for higher spring constants at equal indentation depth stiff inclusions can be detected within the surrounding soft matrix by use of the developedFigure 5 shows the comparison of the results for a blank liver sample and one equipped with a hard silicone node (Shore A40, Ř 7 mm). Both images show sensor data acquired at an applied pressure of 4.0·10−3 N/mm2 which was calculated by use of the weight values and the size of the sensor area. It can clearly be seen that a stiff inclusion was present within the second sample as high values are present in the lower left corner and the middle of the bar graphs.
Moreover, the manually performed tests revealed that seven of the eight subjects localized all the artificial tumors hidden within the liver samples with high confidence as shown in Figure 6. They took 18s to 80s to find a lump and mostly described the handling as “intuitive” or “simple after getting used to”.
Subjects had to figure out the optimal amount of pressure for palpation on their own to achieve a suitable sensor output. However, they all rapidly noticed that too little forces lead to improper or incomplete contact with the examined tissue, thus, to a weak and inexpressive signal. Furthermore, it was immediately recognized by all of the test persons that when forces were chosen too high, artifacts were caused within the visualized output. These “ghost” tumors could easily be excluded by reducing the load and consequently noticing the simultaneous decrease of the sensor signal over the complete area without any bar graphs remaining at high level. It was calculated that they generally applied pressure values between 6.37·10−3 N/mm2 and 6.51·10−3 N/mm2 with a median value of 6.48·10−3 N/mm2. For one of the subjects an extreme value of 8.15·10−3 N/mm2 was noted which is visualized as a cross above the boxplot in Figure 7.
4 Discussion and future work
The results of the automated palpation tests revealed that it was not possible to distinguish different sizes of stiff inclusions. Moreover, very small artificial nodes (Ř 3.5 mm) were not visible at all within the output signal of the sensor. This is due to the fact that the PBCTs of the sensor array are located in a distance of 2.5 mm which means that the resolution of the sensor is not high enough to recognize variations of only a few millimeters and small inclusions are not detected as they may be located in between the sensitive lines. It is important to improve the sensor resolution as especially tumors with a diameter less than 5 mm cannot be visualized by medical imaging .
Trejos et al. [6, 7] stated that with their tactile probe 4N were the optimal amount of force for the detection of hepatic cancer whereas 6N were the maximum load before damage occurred to the delicate tissue. For better comparability the associated pressure values were calculated based on the measurements of their sensitive tool tip. Hence, according to these parameters the optimal pressure is 16.7·10−3 N/mm2 and the maximum 25.0·10−3 N/mm2. Comparing these values to the data acquired during the test series with eight subjects using the instrument with the laterally opening lid it can be said that the new design requires less pressure to achieve suitable sensor signals for tumor detection. More precisely, the median pressure of the subject tests was only 38.8% of the optimal and 25.9% of the maximum determined pressure value. Even the extreme value was only half of the optimal pressure, thus, the applied palpation forces are not critical in terms of causing damage to the examined liver tissue.
It was noticed that artifacts or “ghost” tumors may occur when forces are too high during palpation. Although the test persons did not complain about this fact and were able to distinguish an artifact from a real hard inclusion it should be considered to integrate a warning system for future designs which informs the user about overabundantly applied pressure.
Furthermore, it must be taken into account that currently raw data has been used to create the 3D color-coded bar graphs as well as the real-time visualization of the sensor output for the subject tests. Other groups [1, 2, 8, 9] use filters and image processing to improve the outcome, however, for the experiments with the new laterally opening lid concept it was decided not to affect the acquired data in order to achieve genuine results.
Concerning the mechanical requirements mentioned at the beginning the mechanical structure of the developed instrument has to be enhanced with an articulating head in a future version. The long, slender laparoscopic devices are not easy to handle through the small ports provided by the trocars. Hence, the tool design must incorporate features such as articulating joints to simplify their maneuverability for the surgeon.
It has been shown that the basic mechanical design of the developed prototype can carry a tactile sensor that allows the detection and localization of hard silicone spheres (Shore A40) embedded in a soft silicone matrix and porcine liver specimen. Based on the promising results of this preliminary study a commercially available device can be implemented additionally taking biocompatibility and sterilizability into account. Furthermore, the sensor design must be adapted to the mechanical restrictions, e.g. the wires must run through the hollow shaft of the instrument and none of the sensor components may exceed the edges of the mechanical structure. This is especially important in the closed state of the device as it would otherwise not fit through a 10 mm trocar. Additionally, the tactile tool has to be adapted to the medical environment, e.g. the electronics must be compatible to the human body and vice versa.
Trejos, A. A Sensorized Instrument for Minimally Invasive Surgery for the Measurement of Forces during Training and Surgery: Development and Applications. Dissertation. The University of Western Ontario 2012, 218 p. Google Scholar
Schostek, S. Taktile Sensorik in der laparoskopischen Chirurgie, Systementwicklung und experimentelle Evaluierung. Dissertation. Eberhard-Karls-Universität zu Tübingen 2010, 136 p. Google Scholar
Strohmayr, M.W. Artificial Skin in Robotics - A Comprehensive Interface for System-Environment Interaction. Dissertation. Karlsruher Institut für Technologie (KIT) 2012, 217 p. Google Scholar
Fröhlich, M. Towards realistic haptic organ phantoms for medical training on minimally invasive robotic surgery systems. Diploma thesis. Technical University Munich 2012, 115 p. Google Scholar
Perri, M., Trejos, A., Naish, M., et al. Short Papers System for Minimally Invasive Tumor Localization. IEEE/ASME Transactions on Mechatronics 2010, 15(6), p. 925 – 931 Google Scholar
Trejos, A., Jayender, J., Perri, M., et al. Experimental Evaluation of Robot-Assisted Tactile Sensing for Minimally Invasive Surgery. IEEE/RAS-EMBS International Conference on Biomedical Robotics and Biomechatronics 2008, p. 971-976 Google Scholar
Trejos, A., Jayender, J., Perri, M., et al. Robot-Assisted Tactile Sensing for Minimally Invasive Tumor Localization.The International Journal of Robotics Research 2009, 28(9), p.1118-1133 Google Scholar
Deml, B., Ortmaier, T., Seibold, U. The Touch and Feel in Minimally Invasive Surgery. IEEE International Workshop on Haptic Audio Visual Environments and their Applications, Ottawa 2005, p.33-38 Google Scholar
Miller, A., Peine, W., Son, J., et al. Tactile Imaging System for Localizing Lung Nodules during Video Assisted Thoracoscopic Surgery. IEEE International Conference on Robotics and Automation 2007, p.2996-3001 Google Scholar
About the article
Published Online: 2015-09-12
Published in Print: 2015-09-01
Conflict of interest: Authors state no conflict of interest. Material and Methods: Informed consent: Informed consent has been obtained from all individuals included in this study. Ethical approval: The research related to human use has been complied with all the relevant national regulations, institutional policies and in accordance the tenets of the Helsinki Declaration, and has been approved by the authors’ institutional review board or equivalent committee.