- Email: [email protected]
Design, construction and performance evaluation of the target tissue thickness measurement system in intraoperative radiotherapy for breast cancer
Author’s Accepted Manuscript Design, construction and performance evaluation of the target tissue thickness measurement system in intraoperative radiotherapy for breast cancer Mohammad Reza Yazdani, Saeed Setayeshi, Hossein Arabalibeik, Mohammad Esmaeil Akbari www.elsevier.com/locate/nima
PII: DOI: Reference:
S0168-9002(16)31333-X http://dx.doi.org/10.1016/j.nima.2016.12.065 NIMA59555
To appear in: Nuclear Inst. and Methods in Physics Research, A Received date: 17 May 2016 Revised date: 31 December 2016 Accepted date: 31 December 2016 Cite this article as: Mohammad Reza Yazdani, Saeed Setayeshi, Hossein Arabalibeik and Mohammad Esmaeil Akbari, Design, construction and performance evaluation of the target tissue thickness measurement system in intraoperative radiotherapy for breast cancer, Nuclear Inst. and Methods in Physics Research, A, http://dx.doi.org/10.1016/j.nima.2016.12.065 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Design, construction and performance evaluation of the target tissue thickness measurement system in intraoperative radiotherapy for breast cancer
1
Mohammad Reza Yazdani
4 5 6
Faculty of Energy Engineering and Physics, Amirkabir University of Technology, Tehran, Iran. Email: [email protected]
Saeed Setayeshi Faculty of Energy Engineering and Physics, Amirkabir University of Technology, Tehran, Iran. Tel: +982164545252 Email: [email protected]
Hossein Arabalibeik (corresponding author) Research Center for Biomedical Technology and Robotics (RCBTR), Tehran University of Medical Sciences, Tehran, Iran. Tel: +989122466620 Email: [email protected]
Mohammad Esmaeil Akbari Cancer Research Center, Shahid Beheshti University of Medical Sciences, Tehran, Iran.
2 3
7 8 9 10 11 12 13 14 15 16 17
Abstract
18
Intraoperative electron radiation therapy (IOERT), which uses electron beams for irradiating the target directly during the surgery, has the advantage of delivering a homogeneous dose to a controlled layer of tissue. Since the dose falls off quickly below the target thickness, the underlying normal tissues are spared. In selecting the appropriate electron energy, the accuracy of the target tissue thickness measurement is critical. In contrast to other procedures applied in IOERT, the routine measurement method is considered to be completely traditional and approximate. In this work, a novel mechanism is proposed for measuring the target tissue thickness with an acceptable level of accuracy. An electronic system has been designed and manufactured with the capability of measuring the tissue thickness based on the recorded electron density under the target. The results indicated the possibility of thickness measurement with a maximum error of 2 mm for 91.35 percent of data. Aside from system limitation in estimating the thickness of 5 mm phantom, for 88.94 percent of data, maximum error is 1 mm.
19 20 21 22 23 24 25 26 27 28 29 30 31 32
Keywords: Breast cancer; Intraoperative electron radiotherapy; Target tissue thickness; Measurement system.
33 34 35
1. Introduction
36
Breast cancer is the second leading reason of cancer mortality among women [1,2]. Amongst the various methods of cancer treatment, the most practical and effective approaches are surgery and radiotherapy (RT) respectively [3]. Widespread use of radiotherapy has led to a decrease in breast cancer mortality rates since 1975 [4]. However, there is no general agreement on whether the entire breast requires to be irradiated [5]. The accelerated partial breast irradiation (APBI), based on limiting the irradiation to the proximity of the tumor bed has contributed to alterations in the RT paradigms leading to shortened course of treatment and more patient comfort [6,7]. There is significant interest in improving techniques for partial breast irradiation (PBI) and recognizing patient subgroups that may take advantage of this approach. Proponents of PBI remark on the conformance of several benefits of this treatment over conventional whole breast EBRT including both sparing of normal tissues and having a shorter RT course that could enhance patient comfort and compliance [8,9,10,11,12,13,14].
37
One kind of PBI is called intraoperative radiotherapy (IORT), which is typically delivered in a single fraction at the time of partial mastectomy (PM). Different techniques for breast IORT have been implemented for two decades; however, high-level clinical data have only recently come out from randomized experiments [8,9,15,16,17,18].
52
In breast IORT, a single radiation dose is delivered under direct visual inspection of the tumor bed. It thus diminishes the local recurrence risk and reduces toxicity since a lower radiation dose is delivered to the healthy tissues [19].
57
Radiation damages structure of all cells, but healthy cells can be rehabilitated more than tumor cells. Therefore, tumor cells are damaged relatively more compared to other cells during the radiotherapy [20].
60
Currently, IORT can be performed with the use of either electron or x-ray beams. The advantage of the latter is well manifested in the treatments where the tumor mass is completely resected and consequently the residual cavity becomes entirely available. Therefore, it is possible to irradiate both the suspected regions and tumor bed, homogenously [21].
63
38 39 40 41 42 43 44 45 46 47 48 49 50 51
53 54 55 56
58 59
61 62
64 65 66 67
Intraoperative electron radiation therapy (IOERT) which directly uses electron beams for irradiating the tumor or its bed during the surgery, is mainly helpful for those patients who-suffer a grossly whole tumor resection. It is also useful when it is impossible to resect the tumor mass by surgery. The use of electron beams permits delivering a homogeneous dose to a controlled layer of tissue. The dose falls off quickly below the target thickness, hence the underlying normal issues are spared. Potential side effects associated with conventional radiation therapy, such as irradiation of the skin or lung and heart tissues, can be reduced or completely eliminated [22,23,24].
68
In order to set up the system for irradiating the tissue in breast IORT, it is necessary to determine the maximum and minimum thicknesses of the target. A protective disk is placed under the target and the surgeon inserts a needle into various parts of the target and then using a ruler, measures the value of the needle penetration inside the target tissue. The surgeon tries to approximate minimum thickness of the tissue and report it to the radiotherapist for dose calculation and energy determination [25].
77
The accuracy of this measurement is more critical when electron beams are utilized because of their finite depth of treatment. In contrast to other procedures applied in IORT, which are performed based on accurate calculations and advanced instrumentation, this measurement is completely traditional and approximate.
84
To the best of our knowledge, no research has been conducted for enhancing this conventional way of measurement yet. In this work, a novel operational mechanism is proposed for measuring the target volume thickness with an acceptable accuracy. The main idea is to use electron penetrability in tissue for measuring the target thickness in breast IOERT. By designing an electrical circuit and detecting the crossing electrons, it would be possible to interpret target thickness.
88
An electronic system is designed and manufactured which is capable of measuring the target thickness based on recording density of electrons under the target tissue in breast cancer IOERT.
95
There are several devices based on different methods for electron detection such as Microchannel Plate Detector [26], CCD [27] and Faraday Cup [28]. In this work, we need to measure the flow of a high energy electron beam. Considering the
98
69 70 71 72 73 74 75 76
78 79 80 81 82 83
85 86 87
89 90 91 92 93 94
96 97
99 100
requirement that the measurement should be performed in a limited space, and the fact that there is no need for electron multiplication in high flows, the Faraday Plate [28] would be a reasonable choice.
101
2. Mechanism of the system performance
105
When the patient is ready for irradiation, the designed system is placed on the protective disk underneath the target tissue. Before the main treatment routine, the target is exposed with a measurement radiation in a short period of time. Depending on the tissue thickness over the sensor in each part, different values of electron current cross the tissue bulk and reach the electronic system. The quantity of current induced in each section of the disk could be used as a measure of thickness for the tissue above it. The recorded signal is analyzed in the system and finally the thickness in each part of the tissue is obtained and could be used for calculations of the main irradiation such as energy selection or dose distribution estimation.
106
2.1 Electron energy selection
117
Currently, energies from 3 to 12 MeV is clinically applicable [29,30].To record the required data, the constructed electrical board was placed under each phantom and irradiated with 3, 6, 8, 10 and 12 MeV electron beams utilizing a typical accelerator system (CLINAC 2100C, Varian Medical Systems Inc., United States). According to the test results applied in various energies, 6 MeV electrons created the highest signal intensity difference in various thicknesses. Therefore, 6 MeV electrons with the practical penetration range of about 30 mm in soft tissue [ 31] were applied (Fig.1.).
118
102 103 104
107 108 109 110 111 112 113 114 115 116
119 120 121 122 123 124 125
126 Fig. 1. Recorded system output for different depths and beam energies. Considering the obtained results and analyzing the recorded signal, 6 MeV is recognized as the most appropriate energy for separating different thicknesses.
127 128 129 130
2.2. Printed Circuit Board (PCB) manufacturing
131
For detecting the electron fluence crossing the target tissue thickness, a printed circuit board (PCB) was designed and manufactured. The thin metal layer on the PCB board, senses the electron fluence acting as a Faraday Plate. Each PCB layer is divided into 16 separate cells (Fig. 2, left). Each of the 16 segments is considered a pixel and its received signal is analyzed separately. For amplifying the electrical current and preventing the effect of electron fluence related to each segment on other parts, an independent electrical circuit is designed on the back side of each pixel region (Fig. 2, right).
132 133 134 135 136 137 138 139 140
141
Fig.2. The manufactured PCB. Left: front view; Right: back view.
142
2.3. Signal amplification, Digitization and Data processing
143
Considering the fact that the received electron current after crossing the target thickness is weak (in the range of 10 nA), a transistor circuit is utilized to amplify the signal in three steps. The signal is sampled using a 1 KHz analog to digital convertor and the result is transferred to the computer via a USB port for further processing. A computer software, developed for analyzing the data, is used to filter the received signal using wavelet analysis first, and then to extract appropriate features for the thickness estimation. The process steps are depicted in the block diagram of Figure 3.
144 145 146 147 148 149 150 151
152 Fig.3. Block diagram of the system
153
3. Calibration and the evaluation of system performance
154
3.1. Tissue equivalent phantoms
155
In order to calibrate system and evaluate its performance, tissue equivalent Poly Methyl Methacrylate (PMMA) phantoms with 1.190 g/cm3 density are. To resemble different tissue depths, six phantoms with various thicknesses were
156 157 158
designed and manufactured. All phantoms had circular cross sections with a diameter of 10 cm. Detailed characteristics of phantoms are as follows:
159
1. A cylinder with the height of 5 mm
161
2. A cylinder with the height of 10 mm
162
3. A cylinder with the height of 15 mm
163
4. A cylinder with the height of 20 mm
164
5. A cylinder with the height of 25 mm
165
6. A cylinder with the cross section divided into 4 quadrants with the heights of 5, 10, 15 and 20 mm, respectively (Fig.4).
166
160
167 168 169
170 Fig.4. Tissue equivalent PMMA phantoms with various thicknesses.
171 172
3.2. Signal recording
173
To record the required data, the constructed electrical board was placed under each phantom and irradiated with 6 MeV electron beam. Sixteen various signals, equivalent to 16 separated pixels on the board, were recorded in each irradiation for 10 s. The obtained data from the first round of data collection consisting of two
174 175 176 177
sessions are enough for system calibration and test reproducibility evaluation. To evaluate the system more, a second round of recording was performed and the resulting data was used as test data. Number of the performed tests and recorded signals are indicated in Table 1.
178 179 180 181 182 183
Table 1.
184
Number of the tests done and recorded signals.
185
Type of data
Phantom number
Number of Number of irradiations Number of signal irradiation for each phantom in recording canals in each sessions each session irradiation
Total signal number
Calibration 1, 2, 3, 4, 5
2
3
16
480
1, 2, 3, 4, 5
2
2
16
320
6
4
3
16
192
Test 186 187
3.3. Data analysis
188
The software first filters the signal using wavelet analysis [32]. Then the mean, middle, maximum and minimum values are extracted from each of 480 calibration signals. Mean value of each parameter for each thickness is calculated next. Finally, to extract a mapping between a parameter and the phantom thickness, a polynomial curve is fitted to the parameter average values and their related phantom thicknesses. This leads to 4 estimation curves.
189
To measure a new sample thickness, we use these four curves to estimate four values for its thickness, with their average reported as the final sample thickness.
195
190 191 192 193 194
196 197
4. Results and discussion
198
Figures 5 and 6 indicate typical recorded raw and filtered signals when 5 mm and 15 mm phantoms (phantom numbers 1 and 3) were irradiated by 6 MeV electron beam, respectively. The recorded signal for 5 mm phantom is depicted in Fig. 5
199 200 201
(right), while Fig. 5 (left) shows the filtered signal using level 5 approximation of Daubechies D4 wavelet. Fig. 5 (middle) compares these signals. The related curves for 15 mm case are represented in Fig. 6 (right), Fig. 6 (middle) and Fig.6 (left), respectively.
202 203 204 205 206 207
208 Fig.5. A sample of recorded signal for the phantom with 5mm thickness (left), the separation of low frequency signal (middle), signal after filtration (right).
209 210
211 Fig.6. A sample of recorded signal for phantom with 15mm thickness (left), the separation of low frequency signal (middle) signal after filtration (right).
212 213 214
According to Fig. 5 and Fig. 6, with an increase in the thickness of the phantom, the system output of the signal decreases remarkably. Therefore, by comparing the filtered signals system outputs, it would be possible to recognize differences across phantom thicknesses.
215
Table 2 represents the average of means, middles, maximums and minimums among all filtered signals for each phantom thickness.
219
216 217 218
220
Table 2.
221
The average of the extracted values from the filtered calibration signals.
222
Phantom thickness (mm) 5 10 15 20 25
Average of Minimum values 72.06 66.44 43.13 16.74 6.62
Average of Maximum values 93.05 84.28 57.35 21.12 7.70
Average of Mean values 82.04 74.98 51.58 20.09 7.04
Average of Middle values 78.80 72.35 49.89 19.05 7.16
223
Fig.7 depicts the mapping between average values of signal means (Table 2, column 4) and the phantom thickness (Table 2, column 1) using a 4th degree polynomial (R2=1.0000). The error bar represents the variation of signal mean in calibration measurements for each thickness.
224 225 226 227
100 90 80
System output
70 60 50 40 30 20 10 0 0
5 mean
10
15
Depth(mm)
Max. Background
20 21.89 Poly. (mean)
25
R=1.0000
Fig.7. Mapping between mean system output of recorded signal and phantom thickness. “Max. Background” is the equivalent of recorded signal, when there is no irradiation. The trend line fitted into minimum of error bars crosses the maximum of background in 21.89 mm.
228 229 230 231 232
According to Fig.7, the sensitivity is less in the 10-25 mm range compared to 5-10 mm range. For 25 mm thickness, there is no significant differences between estimations obtained from irradiated and non-irradiated phantoms. We have denoted the supremum of all signals for non-irradiated samples as the background in fig. 7. The error band of the fitted curve crosses the background signal in 21.89 mm thickness which conservatively suggests that the system response is valid up to 21.89 mm thickness. The performance of measurement system for test data are presented in Table 3.
233 234 235 236 237 238 239 240 241
Table 3.
242
System performance for test data.
243
Phantom number
Thickness (mm)
Total # test data
# results without error
# results with 1mm error
# results with 2mm error
# results with 3mm error
# results with 4mm error
1
5
160
94
29
22
12
3
2
10
160
122
20
11
7
0
3
15
160
131
14
11
4
0
4
20
160
127
16
15
2
0
5
25
160
5
48
25
12
8
2
1
10
48
34
9
3
2
0
15
48
34
10
3
1
0
20
48
33
12
2
1
0
tissue thickness is more than 21.89 mm.
6
244
Table 3 indicates the weak performance of the system for 5 mm phantom compared to other phantoms in the range 10-21.89 mm. This is mainly because of the low sensitivity in the range 5-10 mm on one hand, and the loss of data for thicknesses less than 5 mm on the other hand. To compensate for the lack of data, there is a need for more test results in the range 0-5 mm. Since the target thickness is more than 1 cm most of the time, in this work we have intentionally excluded the investigation for thicknesses less than 5 mm. In order to have a more sensible
245 246 247 248 249 250 251
evaluation of the system error, accumulative percentage of the acceptable system responses at various errors are presented in Table 4.
252
Table 4.
254
Accumulative percentage of the acceptable system response for test data at various errors.
255
Thickness (mm) 5 10 15 20
Results without error (%) 57.21 75.00 79.33 76.92
Results with maximum 1mm error (%) 76.92 88.94 90.87 90.38
253
Results with maximum 2mm error (%) 91.35 95.67 97.60 98.56
256
According to the results shown in Table 4, the designed system presents a correct estimation within the 2 mm error band for more than 90% of cases in the worst case (the minimum examined thicknesses). Ignoring the results for 5 mm thickness, the system estimates nearly 89% of cases with a 1 mm error bound in the worst cases. These errors, which are less than 10% in the second scenario, are acceptable and could be ignored compared to the routine methods which are currently used by surgeons.
257
To improve the accuracy of measurement in the low-thickness range, there are two suggestions: using other features of the recorded signal for estimating the thickness and improving the designed circuit to enhance its performance.
264
For recording signals in-vivo, we should take into account the effects of blood and conducting tissue that would touch the board. The first, if not the most important requirement would be to isolate the PCB circuit from the tissue and make it waterproof using an appropriate sealing coat. Furthermore, although we realize that tissue differences have some effects on the final results, we have neglected this effects here as the first step of a comprehensive study.
267
258 259 260 261 262 263
265 266
268 269 270 271 272 273
5. Conclusion
274
The main intention of this research was developing a new mechanism for measuring the target tissue thickness during IOERT for breast cancer. This novel method could omit the human role in thickness estimation and increase the
275 276 277
accuracy of measurement. The obtained results indicate that for 91.35% of the cases, the system measures phantom thickness with a maximum of 2 mm error. Should we ignore the system weakness in estimating 5 mm thickness, the measurement error for 88.94% of data is 1 mm at most.
278
Compared to the traditional methods of measuring the tissue thickness in IORT, the presented method is much more satisfactory. Preliminary results are promising and encourage one to conduct further studies for overcoming the current shortcomings and improving the performance of the system. We hope that this would lead to an applicable commercial tool in manufacturing future IORT systems.
282
Our future work will focus on improving the system performance and increasing the measurement accuracy through:
288
1. Examining the thicknesses less than 5 mm,
290
2. Improving the amplification of the recorded signal for enhancing the system accuracy in recognizing the signal differences,
291
3. Using other features of the signal which are not discussed in this work,
293
4. Increasing the number of pixels on the electrical board which can lead to a higher spatial resolution of measurement.
294
279 280 281
283 284 285 286 287
289
292
295 296
References 1
Stearns, Vered. "Novel Biomarkers in the Continuum of Breast Cancer." (2016). ISBN: 978-3319-22908-9 (Print) 978-3-319-22909-6 (Online). 2
Torre, Lindsey A., et al. "Global cancer statistics, 2012." CA: a cancer journal for clinicians 65.2 (2015): 87-108. 3
DeSantis, Carol E., et al. "Cancer treatment and survivorship statistics, 2014." CA: a cancer journal for clinicians 64.4 (2014): 252-271. 4
Benson, John R., and Katy AT Teo. "Breast cancer local therapy: what is its effect on mortality?." World journal of surgery 36.7 (2012): 1460-1474.
297
5
Huang, Eugene, et al. "Classifying local disease recurrences after breast conservation therapy based on location and histology." Cancer 95.10 (2002): 2059-2067. 6
Cox, John A., and Todd A. Swanson. "Current modalities of accelerated partial breast irradiation." Nature Reviews Clinical Oncology 10.6 (2013): 344-356. 7
Barry, Mitchel, Alice Ho, and Monica Morrow. "The evolving role of partial breast irradiation in early-stage breast cancer." Annals of surgical oncology 20.8 (2013): 2534-2540. 8
Vaidya, Jayant S., et al. "Targeted intraoperative radiotherapy versus whole breast radiotherapy for breast cancer (TARGIT-A trial): an international, prospective, randomised, non-inferiority phase 3 trial." The Lancet 376.9735 (2010): 91-102. 9
Veronesi, Umberto, et al. "Intraoperative radiotherapy versus external radiotherapy for early breast cancer (ELIOT): a randomised controlled equivalence trial." The lancet oncology 14.13 (2013): 1269-1277. 10
Vaidya, Jayant S., et al. "Risk-adapted targeted intraoperative radiotherapy versus wholebreast radiotherapy for breast cancer: 5-year results for local control and overall survival from the TARGIT-A randomised trial." The Lancet 383.9917 (2014): 603-613. 11
Luini, Alberto, et al. "The pilot trial on intraoperative radiotherapy with electrons (ELIOT): update on the results." Breast cancer research and treatment 93.1 (2005): 55-59. 12
Lemanski, Claire, et al. "Electrons for intraoperative radiotherapy in selected breast-cancer patients: late results of the Montpellier phase II trial." Radiation Oncology 8.1 (2013): 1. 13
Lemanski, Claire, et al. "Intraoperative radiotherapy in early-stage breast cancer: results of the montpellier phase II trial." International Journal of Radiation Oncology* Biology* Physics 76.3 (2010): 698-703. 14
Mosalanezhad, Hedie, et al. "Cost-effectiveness of radiotherapy during surgery compared with external radiation therapy in the treatment of women with breast cancer." Journal of Health Management and Informatics 3.2 (2016): 33-38. 15
Shah, Chirag, et al. "Comparison of survival and regional failure between accelerated partial breast irradiation and whole breast irradiation." Brachytherapy 11.4 (2012): 311-315. 16
Veronesi, U., et al. "A preliminary report of intraoperative radiotherapy (IORT) in limitedstage breast cancers that are conservatively treated." European journal of cancer 37.17 (2001): 2178-2183. 17
Polgár, Csaba, et al. "Breast-conserving therapy with partial or whole breast irradiation: tenyear results of the Budapest randomized trial." Radiotherapy and Oncology 108.2 (2013): 197202. 18
Baatjes, K. J., and J. P. Apffelstaedt. "7-year follow up of intra-operative radiotherapy for early breast cancer in a developing country." The Breast 21.3 (2012): 326-329.
19
Mourtada, Firas. "Physics of Intraoperative Radiotherapy for the Breast." Short Course Breast Radiotherapy. Springer International Publishing, 2016. 317-325. 20
Tarnuzzer, Roy W., et al. "Vacancy engineered ceria nanostructures for protection from radiation-induced cellular damage." Nano letters 5.12 (2005): 2573-2577. 21
Gunderson, Leonard L., et al., eds. Intraoperative irradiation: techniques and results. Springer Science & Business Media, 2011. 22
Mourtada, Firas. "Physics of Intraoperative Radiotherapy for the Breast." Short Course Breast Radiotherapy. Springer International Publishing, 2016. 317-325. 23
Mosalanezhad, Hedie, et al. "Cost-effectiveness of radiotherapy during surgery compared with external radiation therapy in the treatment of women with breast cancer." Journal of Health Management and Informatics 3.2 (2016): 33-38. 24
Gunderson, Leonard L., et al., eds. Intraoperative irradiation: techniques and results. Springer Science & Business Media, 2011. 25
Lemanski, Claire, and David Azria. "Intraoperative Technique with Electrons." Short Course Breast Radiotherapy. Springer International Publishing, 2016. 347-359. 26
Blase, Ryan C., et al. "Microchannel Plate Detector Detection Efficiency to Monoenergetic Electrons Between 0.4 and 2.6 MeV." IEEE Transactions on Nuclear Science 62.6 (2015): 33393345. 27
Meyer, Rüdiger R., and Angus I. Kirkland. "Characterisation of the signal and noise transfer of CCD cameras for electron detection." Microscopy research and technique 49.3 (2000): 269280. 28
Eiceman, Gary Alan, Zeev Karpas, and Herbert H. Hill Jr. Ion mobility spectrometry. CRC press, 2013. 29
Yu, Wei, et al. "Intraoperative radiation therapy delivered prior to lumpectomy for early-stage breast cancer: a single institution study." American journal of cancer research 5.7 (2015): 22492257. 30
Orecchia, R., et al. "Intraoperative radiation therapy with electrons (ELIOT) in early-stage breast cancer." The Breast 12.6 (2003): 483-490. 31
AAPM Radiation Therapy Committee, and Faiz M. Khan. Clinical electron-beam dosimetry. American Institute of Physics for the American Association of Physicists in Medicine, 1991. 32
Strang, Gilbert, and Truong Nguyen. Wavelets and filter banks. SIAM, 1996.
PII: DOI: Reference:
S0168-9002(16)31333-X http://dx.doi.org/10.1016/j.nima.2016.12.065 NIMA59555
To appear in: Nuclear Inst. and Methods in Physics Research, A Received date: 17 May 2016 Revised date: 31 December 2016 Accepted date: 31 December 2016 Cite this article as: Mohammad Reza Yazdani, Saeed Setayeshi, Hossein Arabalibeik and Mohammad Esmaeil Akbari, Design, construction and performance evaluation of the target tissue thickness measurement system in intraoperative radiotherapy for breast cancer, Nuclear Inst. and Methods in Physics Research, A, http://dx.doi.org/10.1016/j.nima.2016.12.065 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Design, construction and performance evaluation of the target tissue thickness measurement system in intraoperative radiotherapy for breast cancer
1
Mohammad Reza Yazdani
4 5 6
Faculty of Energy Engineering and Physics, Amirkabir University of Technology, Tehran, Iran. Email: [email protected]
Saeed Setayeshi Faculty of Energy Engineering and Physics, Amirkabir University of Technology, Tehran, Iran. Tel: +982164545252 Email: [email protected]
Hossein Arabalibeik (corresponding author) Research Center for Biomedical Technology and Robotics (RCBTR), Tehran University of Medical Sciences, Tehran, Iran. Tel: +989122466620 Email: [email protected]
Mohammad Esmaeil Akbari Cancer Research Center, Shahid Beheshti University of Medical Sciences, Tehran, Iran.
2 3
7 8 9 10 11 12 13 14 15 16 17
Abstract
18
Intraoperative electron radiation therapy (IOERT), which uses electron beams for irradiating the target directly during the surgery, has the advantage of delivering a homogeneous dose to a controlled layer of tissue. Since the dose falls off quickly below the target thickness, the underlying normal tissues are spared. In selecting the appropriate electron energy, the accuracy of the target tissue thickness measurement is critical. In contrast to other procedures applied in IOERT, the routine measurement method is considered to be completely traditional and approximate. In this work, a novel mechanism is proposed for measuring the target tissue thickness with an acceptable level of accuracy. An electronic system has been designed and manufactured with the capability of measuring the tissue thickness based on the recorded electron density under the target. The results indicated the possibility of thickness measurement with a maximum error of 2 mm for 91.35 percent of data. Aside from system limitation in estimating the thickness of 5 mm phantom, for 88.94 percent of data, maximum error is 1 mm.
19 20 21 22 23 24 25 26 27 28 29 30 31 32
Keywords: Breast cancer; Intraoperative electron radiotherapy; Target tissue thickness; Measurement system.
33 34 35
1. Introduction
36
Breast cancer is the second leading reason of cancer mortality among women [1,2]. Amongst the various methods of cancer treatment, the most practical and effective approaches are surgery and radiotherapy (RT) respectively [3]. Widespread use of radiotherapy has led to a decrease in breast cancer mortality rates since 1975 [4]. However, there is no general agreement on whether the entire breast requires to be irradiated [5]. The accelerated partial breast irradiation (APBI), based on limiting the irradiation to the proximity of the tumor bed has contributed to alterations in the RT paradigms leading to shortened course of treatment and more patient comfort [6,7]. There is significant interest in improving techniques for partial breast irradiation (PBI) and recognizing patient subgroups that may take advantage of this approach. Proponents of PBI remark on the conformance of several benefits of this treatment over conventional whole breast EBRT including both sparing of normal tissues and having a shorter RT course that could enhance patient comfort and compliance [8,9,10,11,12,13,14].
37
One kind of PBI is called intraoperative radiotherapy (IORT), which is typically delivered in a single fraction at the time of partial mastectomy (PM). Different techniques for breast IORT have been implemented for two decades; however, high-level clinical data have only recently come out from randomized experiments [8,9,15,16,17,18].
52
In breast IORT, a single radiation dose is delivered under direct visual inspection of the tumor bed. It thus diminishes the local recurrence risk and reduces toxicity since a lower radiation dose is delivered to the healthy tissues [19].
57
Radiation damages structure of all cells, but healthy cells can be rehabilitated more than tumor cells. Therefore, tumor cells are damaged relatively more compared to other cells during the radiotherapy [20].
60
Currently, IORT can be performed with the use of either electron or x-ray beams. The advantage of the latter is well manifested in the treatments where the tumor mass is completely resected and consequently the residual cavity becomes entirely available. Therefore, it is possible to irradiate both the suspected regions and tumor bed, homogenously [21].
63
38 39 40 41 42 43 44 45 46 47 48 49 50 51
53 54 55 56
58 59
61 62
64 65 66 67
Intraoperative electron radiation therapy (IOERT) which directly uses electron beams for irradiating the tumor or its bed during the surgery, is mainly helpful for those patients who-suffer a grossly whole tumor resection. It is also useful when it is impossible to resect the tumor mass by surgery. The use of electron beams permits delivering a homogeneous dose to a controlled layer of tissue. The dose falls off quickly below the target thickness, hence the underlying normal issues are spared. Potential side effects associated with conventional radiation therapy, such as irradiation of the skin or lung and heart tissues, can be reduced or completely eliminated [22,23,24].
68
In order to set up the system for irradiating the tissue in breast IORT, it is necessary to determine the maximum and minimum thicknesses of the target. A protective disk is placed under the target and the surgeon inserts a needle into various parts of the target and then using a ruler, measures the value of the needle penetration inside the target tissue. The surgeon tries to approximate minimum thickness of the tissue and report it to the radiotherapist for dose calculation and energy determination [25].
77
The accuracy of this measurement is more critical when electron beams are utilized because of their finite depth of treatment. In contrast to other procedures applied in IORT, which are performed based on accurate calculations and advanced instrumentation, this measurement is completely traditional and approximate.
84
To the best of our knowledge, no research has been conducted for enhancing this conventional way of measurement yet. In this work, a novel operational mechanism is proposed for measuring the target volume thickness with an acceptable accuracy. The main idea is to use electron penetrability in tissue for measuring the target thickness in breast IOERT. By designing an electrical circuit and detecting the crossing electrons, it would be possible to interpret target thickness.
88
An electronic system is designed and manufactured which is capable of measuring the target thickness based on recording density of electrons under the target tissue in breast cancer IOERT.
95
There are several devices based on different methods for electron detection such as Microchannel Plate Detector [26], CCD [27] and Faraday Cup [28]. In this work, we need to measure the flow of a high energy electron beam. Considering the
98
69 70 71 72 73 74 75 76
78 79 80 81 82 83
85 86 87
89 90 91 92 93 94
96 97
99 100
requirement that the measurement should be performed in a limited space, and the fact that there is no need for electron multiplication in high flows, the Faraday Plate [28] would be a reasonable choice.
101
2. Mechanism of the system performance
105
When the patient is ready for irradiation, the designed system is placed on the protective disk underneath the target tissue. Before the main treatment routine, the target is exposed with a measurement radiation in a short period of time. Depending on the tissue thickness over the sensor in each part, different values of electron current cross the tissue bulk and reach the electronic system. The quantity of current induced in each section of the disk could be used as a measure of thickness for the tissue above it. The recorded signal is analyzed in the system and finally the thickness in each part of the tissue is obtained and could be used for calculations of the main irradiation such as energy selection or dose distribution estimation.
106
2.1 Electron energy selection
117
Currently, energies from 3 to 12 MeV is clinically applicable [29,30].To record the required data, the constructed electrical board was placed under each phantom and irradiated with 3, 6, 8, 10 and 12 MeV electron beams utilizing a typical accelerator system (CLINAC 2100C, Varian Medical Systems Inc., United States). According to the test results applied in various energies, 6 MeV electrons created the highest signal intensity difference in various thicknesses. Therefore, 6 MeV electrons with the practical penetration range of about 30 mm in soft tissue [ 31] were applied (Fig.1.).
118
102 103 104
107 108 109 110 111 112 113 114 115 116
119 120 121 122 123 124 125
126 Fig. 1. Recorded system output for different depths and beam energies. Considering the obtained results and analyzing the recorded signal, 6 MeV is recognized as the most appropriate energy for separating different thicknesses.
127 128 129 130
2.2. Printed Circuit Board (PCB) manufacturing
131
For detecting the electron fluence crossing the target tissue thickness, a printed circuit board (PCB) was designed and manufactured. The thin metal layer on the PCB board, senses the electron fluence acting as a Faraday Plate. Each PCB layer is divided into 16 separate cells (Fig. 2, left). Each of the 16 segments is considered a pixel and its received signal is analyzed separately. For amplifying the electrical current and preventing the effect of electron fluence related to each segment on other parts, an independent electrical circuit is designed on the back side of each pixel region (Fig. 2, right).
132 133 134 135 136 137 138 139 140
141
Fig.2. The manufactured PCB. Left: front view; Right: back view.
142
2.3. Signal amplification, Digitization and Data processing
143
Considering the fact that the received electron current after crossing the target thickness is weak (in the range of 10 nA), a transistor circuit is utilized to amplify the signal in three steps. The signal is sampled using a 1 KHz analog to digital convertor and the result is transferred to the computer via a USB port for further processing. A computer software, developed for analyzing the data, is used to filter the received signal using wavelet analysis first, and then to extract appropriate features for the thickness estimation. The process steps are depicted in the block diagram of Figure 3.
144 145 146 147 148 149 150 151
152 Fig.3. Block diagram of the system
153
3. Calibration and the evaluation of system performance
154
3.1. Tissue equivalent phantoms
155
In order to calibrate system and evaluate its performance, tissue equivalent Poly Methyl Methacrylate (PMMA) phantoms with 1.190 g/cm3 density are. To resemble different tissue depths, six phantoms with various thicknesses were
156 157 158
designed and manufactured. All phantoms had circular cross sections with a diameter of 10 cm. Detailed characteristics of phantoms are as follows:
159
1. A cylinder with the height of 5 mm
161
2. A cylinder with the height of 10 mm
162
3. A cylinder with the height of 15 mm
163
4. A cylinder with the height of 20 mm
164
5. A cylinder with the height of 25 mm
165
6. A cylinder with the cross section divided into 4 quadrants with the heights of 5, 10, 15 and 20 mm, respectively (Fig.4).
166
160
167 168 169
170 Fig.4. Tissue equivalent PMMA phantoms with various thicknesses.
171 172
3.2. Signal recording
173
To record the required data, the constructed electrical board was placed under each phantom and irradiated with 6 MeV electron beam. Sixteen various signals, equivalent to 16 separated pixels on the board, were recorded in each irradiation for 10 s. The obtained data from the first round of data collection consisting of two
174 175 176 177
sessions are enough for system calibration and test reproducibility evaluation. To evaluate the system more, a second round of recording was performed and the resulting data was used as test data. Number of the performed tests and recorded signals are indicated in Table 1.
178 179 180 181 182 183
Table 1.
184
Number of the tests done and recorded signals.
185
Type of data
Phantom number
Number of Number of irradiations Number of signal irradiation for each phantom in recording canals in each sessions each session irradiation
Total signal number
Calibration 1, 2, 3, 4, 5
2
3
16
480
1, 2, 3, 4, 5
2
2
16
320
6
4
3
16
192
Test 186 187
3.3. Data analysis
188
The software first filters the signal using wavelet analysis [32]. Then the mean, middle, maximum and minimum values are extracted from each of 480 calibration signals. Mean value of each parameter for each thickness is calculated next. Finally, to extract a mapping between a parameter and the phantom thickness, a polynomial curve is fitted to the parameter average values and their related phantom thicknesses. This leads to 4 estimation curves.
189
To measure a new sample thickness, we use these four curves to estimate four values for its thickness, with their average reported as the final sample thickness.
195
190 191 192 193 194
196 197
4. Results and discussion
198
Figures 5 and 6 indicate typical recorded raw and filtered signals when 5 mm and 15 mm phantoms (phantom numbers 1 and 3) were irradiated by 6 MeV electron beam, respectively. The recorded signal for 5 mm phantom is depicted in Fig. 5
199 200 201
(right), while Fig. 5 (left) shows the filtered signal using level 5 approximation of Daubechies D4 wavelet. Fig. 5 (middle) compares these signals. The related curves for 15 mm case are represented in Fig. 6 (right), Fig. 6 (middle) and Fig.6 (left), respectively.
202 203 204 205 206 207
208 Fig.5. A sample of recorded signal for the phantom with 5mm thickness (left), the separation of low frequency signal (middle), signal after filtration (right).
209 210
211 Fig.6. A sample of recorded signal for phantom with 15mm thickness (left), the separation of low frequency signal (middle) signal after filtration (right).
212 213 214
According to Fig. 5 and Fig. 6, with an increase in the thickness of the phantom, the system output of the signal decreases remarkably. Therefore, by comparing the filtered signals system outputs, it would be possible to recognize differences across phantom thicknesses.
215
Table 2 represents the average of means, middles, maximums and minimums among all filtered signals for each phantom thickness.
219
216 217 218
220
Table 2.
221
The average of the extracted values from the filtered calibration signals.
222
Phantom thickness (mm) 5 10 15 20 25
Average of Minimum values 72.06 66.44 43.13 16.74 6.62
Average of Maximum values 93.05 84.28 57.35 21.12 7.70
Average of Mean values 82.04 74.98 51.58 20.09 7.04
Average of Middle values 78.80 72.35 49.89 19.05 7.16
223
Fig.7 depicts the mapping between average values of signal means (Table 2, column 4) and the phantom thickness (Table 2, column 1) using a 4th degree polynomial (R2=1.0000). The error bar represents the variation of signal mean in calibration measurements for each thickness.
224 225 226 227
100 90 80
System output
70 60 50 40 30 20 10 0 0
5 mean
10
15
Depth(mm)
Max. Background
20 21.89 Poly. (mean)
25
R=1.0000
Fig.7. Mapping between mean system output of recorded signal and phantom thickness. “Max. Background” is the equivalent of recorded signal, when there is no irradiation. The trend line fitted into minimum of error bars crosses the maximum of background in 21.89 mm.
228 229 230 231 232
According to Fig.7, the sensitivity is less in the 10-25 mm range compared to 5-10 mm range. For 25 mm thickness, there is no significant differences between estimations obtained from irradiated and non-irradiated phantoms. We have denoted the supremum of all signals for non-irradiated samples as the background in fig. 7. The error band of the fitted curve crosses the background signal in 21.89 mm thickness which conservatively suggests that the system response is valid up to 21.89 mm thickness. The performance of measurement system for test data are presented in Table 3.
233 234 235 236 237 238 239 240 241
Table 3.
242
System performance for test data.
243
Phantom number
Thickness (mm)
Total # test data
# results without error
# results with 1mm error
# results with 2mm error
# results with 3mm error
# results with 4mm error
1
5
160
94
29
22
12
3
2
10
160
122
20
11
7
0
3
15
160
131
14
11
4
0
4
20
160
127
16
15
2
0
5
25
160
5
48
25
12
8
2
1
10
48
34
9
3
2
0
15
48
34
10
3
1
0
20
48
33
12
2
1
0
tissue thickness is more than 21.89 mm.
6
244
Table 3 indicates the weak performance of the system for 5 mm phantom compared to other phantoms in the range 10-21.89 mm. This is mainly because of the low sensitivity in the range 5-10 mm on one hand, and the loss of data for thicknesses less than 5 mm on the other hand. To compensate for the lack of data, there is a need for more test results in the range 0-5 mm. Since the target thickness is more than 1 cm most of the time, in this work we have intentionally excluded the investigation for thicknesses less than 5 mm. In order to have a more sensible
245 246 247 248 249 250 251
evaluation of the system error, accumulative percentage of the acceptable system responses at various errors are presented in Table 4.
252
Table 4.
254
Accumulative percentage of the acceptable system response for test data at various errors.
255
Thickness (mm) 5 10 15 20
Results without error (%) 57.21 75.00 79.33 76.92
Results with maximum 1mm error (%) 76.92 88.94 90.87 90.38
253
Results with maximum 2mm error (%) 91.35 95.67 97.60 98.56
256
According to the results shown in Table 4, the designed system presents a correct estimation within the 2 mm error band for more than 90% of cases in the worst case (the minimum examined thicknesses). Ignoring the results for 5 mm thickness, the system estimates nearly 89% of cases with a 1 mm error bound in the worst cases. These errors, which are less than 10% in the second scenario, are acceptable and could be ignored compared to the routine methods which are currently used by surgeons.
257
To improve the accuracy of measurement in the low-thickness range, there are two suggestions: using other features of the recorded signal for estimating the thickness and improving the designed circuit to enhance its performance.
264
For recording signals in-vivo, we should take into account the effects of blood and conducting tissue that would touch the board. The first, if not the most important requirement would be to isolate the PCB circuit from the tissue and make it waterproof using an appropriate sealing coat. Furthermore, although we realize that tissue differences have some effects on the final results, we have neglected this effects here as the first step of a comprehensive study.
267
258 259 260 261 262 263
265 266
268 269 270 271 272 273
5. Conclusion
274
The main intention of this research was developing a new mechanism for measuring the target tissue thickness during IOERT for breast cancer. This novel method could omit the human role in thickness estimation and increase the
275 276 277
accuracy of measurement. The obtained results indicate that for 91.35% of the cases, the system measures phantom thickness with a maximum of 2 mm error. Should we ignore the system weakness in estimating 5 mm thickness, the measurement error for 88.94% of data is 1 mm at most.
278
Compared to the traditional methods of measuring the tissue thickness in IORT, the presented method is much more satisfactory. Preliminary results are promising and encourage one to conduct further studies for overcoming the current shortcomings and improving the performance of the system. We hope that this would lead to an applicable commercial tool in manufacturing future IORT systems.
282
Our future work will focus on improving the system performance and increasing the measurement accuracy through:
288
1. Examining the thicknesses less than 5 mm,
290
2. Improving the amplification of the recorded signal for enhancing the system accuracy in recognizing the signal differences,
291
3. Using other features of the signal which are not discussed in this work,
293
4. Increasing the number of pixels on the electrical board which can lead to a higher spatial resolution of measurement.
294
279 280 281
283 284 285 286 287
289
292
295 296
References 1
Stearns, Vered. "Novel Biomarkers in the Continuum of Breast Cancer." (2016). ISBN: 978-3319-22908-9 (Print) 978-3-319-22909-6 (Online). 2
Torre, Lindsey A., et al. "Global cancer statistics, 2012." CA: a cancer journal for clinicians 65.2 (2015): 87-108. 3
DeSantis, Carol E., et al. "Cancer treatment and survivorship statistics, 2014." CA: a cancer journal for clinicians 64.4 (2014): 252-271. 4
Benson, John R., and Katy AT Teo. "Breast cancer local therapy: what is its effect on mortality?." World journal of surgery 36.7 (2012): 1460-1474.
297
5
Huang, Eugene, et al. "Classifying local disease recurrences after breast conservation therapy based on location and histology." Cancer 95.10 (2002): 2059-2067. 6
Cox, John A., and Todd A. Swanson. "Current modalities of accelerated partial breast irradiation." Nature Reviews Clinical Oncology 10.6 (2013): 344-356. 7
Barry, Mitchel, Alice Ho, and Monica Morrow. "The evolving role of partial breast irradiation in early-stage breast cancer." Annals of surgical oncology 20.8 (2013): 2534-2540. 8
Vaidya, Jayant S., et al. "Targeted intraoperative radiotherapy versus whole breast radiotherapy for breast cancer (TARGIT-A trial): an international, prospective, randomised, non-inferiority phase 3 trial." The Lancet 376.9735 (2010): 91-102. 9
Veronesi, Umberto, et al. "Intraoperative radiotherapy versus external radiotherapy for early breast cancer (ELIOT): a randomised controlled equivalence trial." The lancet oncology 14.13 (2013): 1269-1277. 10
Vaidya, Jayant S., et al. "Risk-adapted targeted intraoperative radiotherapy versus wholebreast radiotherapy for breast cancer: 5-year results for local control and overall survival from the TARGIT-A randomised trial." The Lancet 383.9917 (2014): 603-613. 11
Luini, Alberto, et al. "The pilot trial on intraoperative radiotherapy with electrons (ELIOT): update on the results." Breast cancer research and treatment 93.1 (2005): 55-59. 12
Lemanski, Claire, et al. "Electrons for intraoperative radiotherapy in selected breast-cancer patients: late results of the Montpellier phase II trial." Radiation Oncology 8.1 (2013): 1. 13
Lemanski, Claire, et al. "Intraoperative radiotherapy in early-stage breast cancer: results of the montpellier phase II trial." International Journal of Radiation Oncology* Biology* Physics 76.3 (2010): 698-703. 14
Mosalanezhad, Hedie, et al. "Cost-effectiveness of radiotherapy during surgery compared with external radiation therapy in the treatment of women with breast cancer." Journal of Health Management and Informatics 3.2 (2016): 33-38. 15
Shah, Chirag, et al. "Comparison of survival and regional failure between accelerated partial breast irradiation and whole breast irradiation." Brachytherapy 11.4 (2012): 311-315. 16
Veronesi, U., et al. "A preliminary report of intraoperative radiotherapy (IORT) in limitedstage breast cancers that are conservatively treated." European journal of cancer 37.17 (2001): 2178-2183. 17
Polgár, Csaba, et al. "Breast-conserving therapy with partial or whole breast irradiation: tenyear results of the Budapest randomized trial." Radiotherapy and Oncology 108.2 (2013): 197202. 18
Baatjes, K. J., and J. P. Apffelstaedt. "7-year follow up of intra-operative radiotherapy for early breast cancer in a developing country." The Breast 21.3 (2012): 326-329.
19
Mourtada, Firas. "Physics of Intraoperative Radiotherapy for the Breast." Short Course Breast Radiotherapy. Springer International Publishing, 2016. 317-325. 20
Tarnuzzer, Roy W., et al. "Vacancy engineered ceria nanostructures for protection from radiation-induced cellular damage." Nano letters 5.12 (2005): 2573-2577. 21
Gunderson, Leonard L., et al., eds. Intraoperative irradiation: techniques and results. Springer Science & Business Media, 2011. 22
Mourtada, Firas. "Physics of Intraoperative Radiotherapy for the Breast." Short Course Breast Radiotherapy. Springer International Publishing, 2016. 317-325. 23
Mosalanezhad, Hedie, et al. "Cost-effectiveness of radiotherapy during surgery compared with external radiation therapy in the treatment of women with breast cancer." Journal of Health Management and Informatics 3.2 (2016): 33-38. 24
Gunderson, Leonard L., et al., eds. Intraoperative irradiation: techniques and results. Springer Science & Business Media, 2011. 25
Lemanski, Claire, and David Azria. "Intraoperative Technique with Electrons." Short Course Breast Radiotherapy. Springer International Publishing, 2016. 347-359. 26
Blase, Ryan C., et al. "Microchannel Plate Detector Detection Efficiency to Monoenergetic Electrons Between 0.4 and 2.6 MeV." IEEE Transactions on Nuclear Science 62.6 (2015): 33393345. 27
Meyer, Rüdiger R., and Angus I. Kirkland. "Characterisation of the signal and noise transfer of CCD cameras for electron detection." Microscopy research and technique 49.3 (2000): 269280. 28
Eiceman, Gary Alan, Zeev Karpas, and Herbert H. Hill Jr. Ion mobility spectrometry. CRC press, 2013. 29
Yu, Wei, et al. "Intraoperative radiation therapy delivered prior to lumpectomy for early-stage breast cancer: a single institution study." American journal of cancer research 5.7 (2015): 22492257. 30
Orecchia, R., et al. "Intraoperative radiation therapy with electrons (ELIOT) in early-stage breast cancer." The Breast 12.6 (2003): 483-490. 31
AAPM Radiation Therapy Committee, and Faiz M. Khan. Clinical electron-beam dosimetry. American Institute of Physics for the American Association of Physicists in Medicine, 1991. 32
Strang, Gilbert, and Truong Nguyen. Wavelets and filter banks. SIAM, 1996.