HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Toshiya Ohtsuka Akira Furuse Tadasu Kohno Jun Nakajima Kuniyoshi Yagyu Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ohtsuka, T. Articles by Omata, S. Search for Related Content PubMed PubMed Citation Articles by Ohtsuka, T. Articles by Omata, S. Related Collections Related Article

Ann Thorac Surg 1995;60:610-613
© 1995 The Society of Thoracic Surgeons

---

Original Articles: General Thoracic

Application of a New Tactile Sensor to Thoracoscopic Surgery: Experimental and Clinical Study

Department of Cardiothoracic Surgery, Faculty of Medicine, University of Tokyo, Tokyo, Japan

	Abstract

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
Background. We developed a new tactile sensor that could quantify the hardness of objects as changes in the resonance frequency of the sensor (f). We have applied it to thoracoscopic operations for the localization of small invisible nodules in the lung.

Methods. When the sensor probe was moved over the lung surface, a f curve was depicted on the computer screen. When the sensor tip reached a point directly above a hard object, a sudden upward jump of the f curve was evoked. After experimental studies using pigs, the sensor was applied in 8 patients. More recently we produced a needle sensor to distinguish small nodules from bronchi that may evoke similar upward jumps of the f curve. Eight nodules and four bronchi in resected human lungs were probed directly using this sensor.

Results. In all of the patients, the hardness of various thoracic structures could be quantified. A total of 10 nodules were found using the sensor and resected thoracoscopically. The needle sensor distinguished nodules from bronchi, as the mean f of the bronchial walls (-64 ± 45.9 Hz) was significantly higher than that of nodules (-526 ± 168 Hz, p < 0.001).

Conclusions. Thoracoscopic detection of small and invisible pulmonary nodules using our new tactile sensor is feasible.

	Introduction

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
See also page 614.

Recently, in many institutions, partial resection of pulmonary tissue to extirpate metastatic tumors or preoperatively indeterminate nodules has been performed routinely using thoracoscopic procedures. With thoracoscopic operations, however, it is extremely difficult to detect precisely small invisible lung nodules located some depth from the lung surface. If such nodules cannot be detected, the procedure has to be converted to open thoracotomy. The reason for such failure is solely that we cannot feel the hardness of some nodules directly with our fingers. To locate invisible nodules through a thoracoscope, we developed a new tactile sensor. The purpose of this communication is to report our experimental and clinical experience with this new tactile sensor.

	Material and Methods

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
Tactile Sensor
PRINCIPLE.
The principle of our new tactile sensor, which has been described elsewhere [1], is as follows. Each material has its own resonance frequency and when the material touches an object vibrating in this frequency, a shift of the resonance frequency is observed. The difference between the frequencies under nontouching and touching conditions (f) depends on the hardness of the object. On the basis of this principle, we produced a new tactile sensor and succeeded in representing the hardness of various thoracic and pulmonary structures by f.

MEASUREMENT SYSTEM.
The sensor measuring system comprises a sensor probe, an amplifier, a filter (patent pending, 1994) and a frequency counter (TR5822, Advantest, Tokyo, Japan) (Fig 1). We have produced a touch sensor probe designed for thoracoscopic operations, which is equipped with a small round tip that is connected acoustically to the piezoelectric transducer made of lead zirconate-barium titanate ceramics. The resonance frequency of this transducer is 92 kHz under nontouching conditions. When the sensor probe touches an object and the resonance frequency shifts, the vibration detector picks up the f and sends a signal to the amplifier that keeps the piezoelectric transducer vibrating in the changed frequency. The f values are processed sequentially by a personal computer (Macintosh PowerBook 145B, Apple Japan Inc, Tokyo, Japan).

View larger version (36K): [in this window] [in a new window]   Fig 1. . The tactile sensor measurement system, which comprises a sensor probe, an amplifier, a filter, and a frequency counter. The sensor probe can be introduced easily into the chest cavity through a 10-mm diameter trocar. The f values are processed swiftly and sequentially by a personal computer (GP-IB = general purpose–interface bass).

View larger version (20K): [in this window] [in a new window]   Fig 2. . Sudden upward jump of a f curve. As the sensor probe was moved over the lung surface (indicated by the broken line), a sudden upward jump (wide arrow) of the f curve occurred when the probe passed within 15 mm of an object harder than the adjacent tissue.

Clinical Application
Since August 1994, in clinical studies, we have used a touch sensor, introduced thoracoscopically, to detect 10 nodules (six indeterminate and four metastatic) in 8 consecutive patients undergoing partial thoracoscopic resection of the lung (Table 1). All nodules detected, except one, were demonstrated by chest computed tomography performed within 1 month before the operation. The diameters of the nodules ranged from 2 to 15 mm and their depths from the chest wall, measured by preoperative chest computed tomography, ranged from 1 to 25 mm (Fig 3). Histologic identification of the six indeterminate nodules had been unsuccessful through a bronchoscope. This is the reason why the thoracoscopic biopsy was applied in these unidentified nodules. All 10 nodules were searched for through a thoracoscope using the technique described for the experimental studies. The point above each target nodule at which the f curve jumped upward was marked with a stitch and partial resection of the lung using a thoracoscopic procedure.

View larger version (133K): [in this window] [in a new window]   Fig 3. . Preoperative chest computed tomography of the nodule 8 showing that the nodule was 3 mm in diameter and 6 mm deep (arrow), which was localized thoracoscopically observing a sudden upward jump of 80 Hz in the f curve.

	Results

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
Animal Experiment
Simulated nodules of 6 and 10 mm in diameter up to 10 mm deep evoked clear sudden upward jumps of the f curves. The frequency of the jump diminished when the nodule was deeply located. Although the smallest nodule (3 mm in diameter) evoked the smallest jump, it could be detected at depth of 8 mm.

Clinical Application
The f values of various human thoracic structures determined were: -896 ± 21 Hz for the esophagus, -818 ± 19 Hz for the superior vena cava, -794 ± 30 Hz for the collapsed lung tissue, -432 ± 72 Hz for the descending aortic wall, and -81 ± 30 Hz for the trachea (Fig 4). The lung surface above three nodules was slightly convex and flat above seven nodules. Imperfect collapse of the lung occurred in 3 patients due to minor intubation problems. Regardless of the flat surface or imperfect collapse in some patients, each nodule evoked a sudden upward jump of the f curve ranging in frequency from 60 to 250 Hz (Table 2). It should be noted that the sixth tumor, the smallest of all, was not revealed by preoperative chest computed tomography. Nevertheless, it was detected by the sensor just beside the seventh tumor. All the nodules were resected successfully thoracoscopically. The histologic diagnoses of the six preoperatively indeterminate nodules were made during the operation (3 tuberculomas, 2 hamartomas, 1 adenocarcinoma). Calcification in each nodule was proved to be absent microscopically. The depth of each nodule from the pleura, measured in the resected specimen in a completely collapsed condition, ranged from 1 to 8 mm.

View larger version (18K): [in this window] [in a new window]   Fig 4. . The mean f values for various human thoracic structures during operation. The mean f values for the abnormal nodules and bronchi in the lungs were obtained using a needle sensor, which proved that the bronchi were significantly harder than the nodules (p < 0.001). (SVC = superior vena cava.)

	Comment

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
The deeper lung nodules are located and the smaller they are, the more difficult it is to find them through thoracoscopy. Forceps cannot be used as an alternative to fingers to detect small nodules because they convey virtually no tactile feeling to the hands. Needle localization and ultrasonography have insurmountable disadvantages [2–6]. Needle localization may cause a hemothorax or pneumothorax and chest computed tomography immediately before intervention is indispensable for this technique. Ultrasonography requires the lung to be completely collapsed to identify a nodule, which cannot be always achieved. Therefore, it is important to develop an instrument that could locate easily and safely small invisible nodules even with residual air in the lung. One day, the senior author (AF) read a newspaper article about a new tactile sensor that could be used to quantify the hardness of the objects. The sensor was originally manufactured by one of the co-authors (SO) to provide robot's fingers with tactile feeling like a human hand. This sensor has been used clinically to evaluate impotence by measuring the stiffness of the penis. We hit upon the idea that as the sensor possessed human hand-like feeling, it might also work well as a tumor detector in thoracoscopic operations. Therefore, we developed a touch sensor specifically tuned for thoracoscopic intervention. Although, the sensor has similar properties to ultrasonography with respect to the application of the sound waves, these two techniques differ considerably. The most distinctive difference is that the sensor takes advantage of f to locate the nodule, not to image it. This is one reason why a small amount of residual air in the lung has a negligible effect on its detection performance. Another reason is that the sensor probe can be pressed against the lung surface safely and approach close to the target. In clinical studies, 10 nodules in 8 patients were found using the sensor and successfully resected through thoracoscopy. It was possible to localize the three nodules under slightly convex surfaces of the lung palpating them with endoscopic forceps. However, localization of the nodules was actually ensured with the sensor observing a sudden upward jump evoked in the f curve. In each patient, a touch sensor was operated easily and safely without injuring the lung tissue, unlike the needle localization. However, we were often asked how we could distinguish small nodules from thin bronchi that might evoke similar and confusing upward jumps of the f curve. Although, in our clinical series, we had not actually experienced any problems discriminating between the two, we produced a needle sensor to solve this potential problem.

Therefore, successful detection of small invisible nodules can be achieved as follows. The touch sensor is used to probe areas in which target nodules have been demonstrated by previous chest computed tomography. If other unidentified objects that evoke similar sudden upward jumps of the f curves, such as bronchi, are present, a needle sensor can be used to differentiate their hardness properties. Objects with f values of around -60 Hz will be bronchi and will be able to be ruled out.

Our new tactile sensor has proved to be a valuable device for detecting pulmonary tumors. However, there are still a few problems concerning its sensory performance. First, the present touch sensor works poorly on the rough surface, such as the lung surface after dissection of adhesion. If the sensor moves jerkily over a rough surface, virtually no reliable f curves are depicted on the screen. A new sensor probe that adjusts to a rough surface, such as, a special probe with a movable head, may solve this problem. Second, as only the pointed tip of the prototype sensor has the sensory ability, scanning a wide area is extremely time consuming. Therefore, we are now producing a new compact sensor with several tips arranged in lines, which enables simultaneous f curves at several points to be recorded.

Furthermore, in clinical studies, we have been collecting f values of various thoracic structures, such as the lung, aorta, and esophagus. We are sure this will be useful for the quantification of fibrotic or sclerotic changes in various organs, such as lung fibrosis and arteriosclerosis.

In conclusion, small invisible nodules that cannot be detected from the lung surface in patients undergoing thoracoscopic operation were located successfully using a new tactile sensor, which is easy to operate, safe to use, and reliable for detection.

	Footnotes

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
Presented at the Thirty-first Annual Meeting of The Society of Thoracic Surgeons, Palm Springs, CA, Jan 30–Feb 1, 1995.

Address reprint requests to Dr Ohtsuka, Department of Cardiothoracic Surgery, Faculty of Medicine, University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo, 113, Japan.

	References

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 

1. Omata S, Terunuma Y. New tactile sensor like the human hand and its applications. Sensors and Actuators 1992;35: 9–15.
2. Shennib H. Intraoperative localization techniques for pulmonary nodules. Ann Thorac Surg 1993;56:745–8.[Abstract]
3. Shepard J, Mathisen D, Muse V, Bhalla M, McLoud T. Needle localization of peripheral lung nodules for video-assisted thoracoscopic surgery. Chest 1994;105:1559–63.[Abstract/Free Full Text]
4. Mack M, Shennib H, Landreneau R, Hazelrigg S. Techniques for localization of pulmonary nodules for thoracoscopic resection. J Thorac Cardiovasc Surg 1993;106:550–3.[Abstract]
5. Dowling R, Keenan R, Ferson P, Landreneau R. Video-assisted thoracoscopic resection of pulmonary metastases. Ann Thorac Surg 1993;56:772–5.[Abstract]
6. Shennib H, Bret P. Intraoperative transthoracic ultrasonographic localization of occult lung lesions. Ann Thorac Surg 1993;55:767–9.[Abstract]

This article has been cited by other articles:

				N. Lizza, P. Eucher, J.-P. Haxhe, J.-F. De Wispelaere, P. M. Johnson, and L. Delaunois Thoracoscopic resection of pulmonary nodules after computed tomographic-guided coil labeling Ann. Thorac. Surg., March 1, 2001; 71(3): 986 - 988. [Abstract] [Full Text] [PDF]

				K. Miyaji, S. Sugiura, H. Inaba, S. Takamoto, and S. Omata Myocardial tactile stiffness during acute reduction of coronary blood flow Ann. Thorac. Surg., January 1, 2000; 69(1): 151 - 155. [Abstract] [Full Text] [PDF]

				M. Tsuchida, Y. Yamato, T. Aoki, T. Watanabe, N. Koizumi, I. Emura, and J.-i. Hayashi CT-Guided Agar Marking for Localization of Nonpalpable Peripheral Pulmonary Lesions Chest, July 1, 1999; 116(1): 139 - 143. [Abstract] [Full Text] [PDF]

				E. D. Blume, K. Altmann, J. E. Mayer, S. D. Colan, K. Gauvreau, and T. Geva Evolution of risk factors influencing early mortality of the arterial switch operation J. Am. Coll. Cardiol., May 1, 1999; 33(6): 1702 - 1709. [Abstract] [Full Text] [PDF]

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Toshiya Ohtsuka Akira Furuse Tadasu Kohno Jun Nakajima Kuniyoshi Yagyu Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ohtsuka, T. Articles by Omata, S. Search for Related Content PubMed PubMed Citation Articles by Ohtsuka, T. Articles by Omata, S. Related Collections Related Article