- Email: [email protected]
Using smartphone as a motion detector to collect time-microenvironment data for estimating the inhalation dose
Author’s Accepted Manuscript Using Smartphone as a Motion Detector to Collect Time-microenvironment Data for estimating the inhalation dose Tran Xuan Hoi, Huynh Truc Phuong, Nguyen Van Hung www.elsevier.com/locate/apradiso
PII: DOI: Reference:
S0969-8043(16)30293-7 http://dx.doi.org/10.1016/j.apradiso.2016.06.024 ARI7529
To appear in: Applied Radiation and Isotopes Received date: 13 January 2016 Revised date: 10 June 2016 Accepted date: 20 June 2016 Cite this article as: Tran Xuan Hoi, Huynh Truc Phuong and Nguyen Van Hung, Using Smartphone as a Motion Detector to Collect Time-microenvironment Data for estimating the inhalation dose, Applied Radiation and Isotopes, http://dx.doi.org/10.1016/j.apradiso.2016.06.024 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.
Using Smartphone as a Motion Detector to Collect Time-microenvironment Data for estimating the inhalation dose Tran Xuan Hoi1,2,*, Huynh Truc Phuong2, Nguyen Van Hung3 1
Faculty of Natural Science, Phu Yen University, 18 Tran Phu―Tuy Hoa, Viet Nam.
2
Faculty of Physics and Engineering Physics, VNUHCM–University of Science, 227 Nguyen Van Cu―Ho Chi Minh City, Viet Nam. 3
Training Center, Nuclear Research Institute, 01 Nguyen Tu Luc―Da Lat, Viet Nam.
[email protected] [email protected] [email protected] *
Corresponding author.
ABSTRACT During the production of iodine-131 from neutron irradiated tellurium dioxide by the dry distillation, a considerable amount of 131 I vapor is dispersed to the indoor air. People who routinely work at the production area may result in a significant risk of exposure to chronic intake by inhaled 131I. This study aims to estimate the inhalation dose for individuals manipulating the 131I at a radioisotope production. By using an application installed on smartphones, we collected the time–microenvironment data spent by a radiation group during work days in 2015. Simultaneously, we used a portable air sampler combined with radioiodine cartridges for grabbing the indoor air samples and then the daily averaged 131I concentration was calculated. Finally, the timemicroenvironment data jointed with the concentration to estimate the inhalation dose for the workers. The result showed that most of the workers had the annual internal dose in 16 mSv. We concluded that using smartphone as a motion detector is a possible and reliable way instead of the questionnaires, diary or GPS-based method. It is, however, only suitable for monitoring on fixed indoor environments and limited the targeted people. Keywords: Iodine-131; Motion detector; time-microenvironment; inhalation dose
1. Introduction Radioiodine-131 is most commonly used for diagnostics and therapy in medicine. People who are occupationally exposed to an internal uptake of iodine includes medical staffs working at nuclear medicine departments and radiation workers of radioisotopes productions. Routine handling of solutions containing radioiodine may result in a significant risk of exposure of the subjects to chronic intake by inhalation of aerosols (IAEA 1999; Bitar et al. 2013; Carneiro et al. 2015; Vidal et al. 2007; Krajewska and Pachocki 2013). A published paper showed that the annual internal effective dose for some workers were above 1 mSv and a worker reached 7.66 mSv with high-risk classification; and this worker must be monitored individually (Bitar et al. 2013). One of the situations needed to be monitored for internal exposure is handling of large quantities of radiopharmaceuticals, such as 131I for therapy (IAEA 1999). The choice of monitoring method depends on factors such as the availability of instrumentation, the analyzes costs, as well as on the sensitivity that is needed. Two in vitro methods were usually used aiming to assess the internal doses due to inhalation intake of 131I: first way was based on the measurement of 131I activity in 24-hour urine samples, and the second one was based on workplace monitoring the 131I aerosol concentration of the indoor air (IAEA 1999). To model exposure to airborne elements, one uses the conceptually simple approach of matching the locations that each exposed person visits with the time-averaged or dynamic air pollutant concentrations that are thought to exist in each visited location (Klepeis 2006; Ott 1982). For obtaining the time people spent in indoor visited locations, researchers used some approaches such as Questionnaire, diaries or GPS-based method (Goldin et al. 2014; Steinle et al. 2013; Carneiro et al. 2015; Glasgow et al. 2014). Diary was usually used as a conventional tool for construct the time-activity pattern. A limiting factor, however, lay in the diaries that only provided restricted information about indoor environments and personal behavior, so the model produces uncertain values in consequence. The complexity and accuracy of the indoor model are limited by the diary accuracy (Gerharz et al. 2009). Using GPS-based method may give an undesirable location precision. GPS-generated time-activity has been well tested and provides an average precision of 7 m in typical urban conditions (Nethery et al. 2014). The precision of 7 m may not be suitable for individual exposure monitoring in normal dimension rooms. In addition, the dependence of GPS-accuracy on many factors such as unfavorable or clustered satellite positions, or ionospheric disturbances in a particular local area is the obstacle must be considered. Another major cause of error is obstruction of the satellite signal. Therefore, GPS-errors will place a participant at the wrong location (Beekhuizen et al. 2013). All the models mentioned above are not including the exact geographical position of the individuals. Researchers who used these methods have to minimize the inconvenience to routine work as well as would get the low accuracy of time-microenvironment (time-ME) data in case of indoor environment monitoring. To avoid these disadvantages, some authors advised that using mobile phone with the help of its application software in tracking environment. Smartphone with inbuilt camera can be used for monitoring fixed places and could lead to more detailed dataset (Broich et al. 2012; Gerharz et al. 2009). Alongside, static air samplers may be used to determine the concentration of airborne radioactive material, which can be combined with site specific assumptions about the physicochemical form of the material and the breathing rate and exposure time of the worker to estimate inhalation intakes (IAEA 1999). In this research, we set the Symbian-based smartphone as a motion detector by installing a commercial application named MotionRecorder which is developed by Ton Nam Software (Ton Nam 2013). This monitoring system uses the built-in camera to detect movements in the surrounding area using an advanced motion detection algorithm without activating the built-in GPS. The application also indicates real-time on the phones display where the movement is detected. The aim of this paper is using low-cost and unwearable devices including a mobile phone’s motion detection application and portable air sampler for estimating individual inhalation dose of workers processing 131I production. 2. Materials and Methods The inhalation dose of workers during the 131I processing period was estimated by using the methodology described in Fig. 1. Fig. 1. Diagram of the methodology. Diary is added when missing data from the Motion Detector, the air sample was grabbed in 2 hr intervals in a particular ME. Daily Individual internal dose was calculated from the daily average of 131I concentration in each ME jointed with time-ME. 1.1. The studied MEs
The study was carried out at the radioiodine production area of Da Lat Nuclear Research Institute in Viet Nam which consists of three rooms (Fig. 2): 131I distilling room (room 1), activity dividing room (room 2) and hot cell room (room 3). Each of these MEs has the same dimension of 6m(L)x5m(W)x4m(H) without any window. The three rooms are connected by two internal doors with room 1 and workers can enter the rooms from an enclosed corridor. Exhaust and air supply systems run continuously during the work time. These MEs are indoor environments and not large. Therefore, in this study, it is assumed that the air concentration in each room is uniform (IAEA 1999). Fig. 2. Experimental arrangement of motion detectors located next to doors in interesting rooms in order to minimize the captured view. Two smartphones needed to monitor indoor exposure of the workers in three rooms, (a) in room 3, (b, c) in room 2. The clock on these phones was updated via Wi-Fi. 1.2. Time-ME collection In order to record time-ME spent by each targeted worker, we used a motion-sensitive application installed in Symbian-based smartphones, named MotionRecorder with light size on disk after installed (Ton Nam 2013). Table 1 shows some characteristics of the “app”. Table 1. MotionRecorder version 1.0 property Items Operating System requirement Size on disk after installed Motion detection sensitivity level available Pixel change sensitivity level available Video resolution output selection Record after movement detected Time-stamp file format Run in background Record audio/mute
Parameter Symbian Anna/Belle 76 kB 0100 0100 HD (1200720) maximum 010 sec *.srt -
Option Adjustable Adjustable Adjustable Adjustable Fixed Available Available
The application turns the phones into a motion-activated video camera. It is run based on the change of two parameters: pixel change sensitivity and motion detection sensitivity. The software allows user can adjust these parameters manually in order to record the motion as expected, and enable smartphone running in background mode to reduce energy usage. Nevertheless, to make sure the smartphone not being in the energy shortage, it was connected to a battery storage (Fig. 2) so that this system could live during a work day. As running, all the movement happening in front of the phone is recorded to a video file subtitled with time and date. When workers enter and leave the monitored rooms, the software records the motion and attaches the real time in hh:mm:ss as soon as these motions occur (Fig. 3). In this research, we used two smartphones in order to monitor the motion occurred at the four doors, each of them is responsible for observing the events at two doors next to each other. The monitor was fixed to the wall next to the doors to confirm that the monitored view was minimized (Fig. 2). The time data and captured video collected by smartphones are manually translated into Microsoft Excel ® file in order to process consequently. Fig. 3. Smartphone installed the Motion Recorder application used as a motion detector to collect time-ME data, (a) the application on standby mode and (b) data file shows timestamp attached to each captured scene. 1.3. Air sampling experimental arrangement Experimental arrangement was performed in the workplace with respect to minimizing the inconvenience to routine work. In regularly, the time of 1.5 hour needed to run a distillation batch in an oven at room 1. At room 2, the group starts to work at noon. People rarely occupied room 3 or the duration is short. A portable air sampler that is Eberline Model RAS-1 and the activated carbon cartridges impregnated with TEDA (Hi-Q Model TC-12) were used to sample the indoor air. The cartridge holder was settled at several position in the rooms near the breathing zone at 1.5m height from the floor (IAEA 2000). The sampling time was about 5 minutes per sample at the flow rate of 70 L/min. The cartridges were sampled in the duration of 510 min corresponding the volume of 0.350.70 m3. 1.4. Determination of 131I in cartridges The cartridges containing 131I were counted at DNRI by gamma spectrometry with a Canberra HPGe coaxial detector system supplied by Oxford Instruments Inc. Nuclear Measurement Group (Model No. CPVDS 30-30185) operated with the pulse height analyzer software Oxford PCA multiport 16K.
Based on some published papers (Montgomery 1990; Kravchik et al. 2008), when relatively small volumes of air are being sampled, it is assumed that most radioiodine is concentrated in the front side of the cassette, greater than 95% of the radioiodine in charcoal cartridge samples was within the front 5 mm and average depth was about 2 mm. Therefore, the detection system used in this counting was calibrated for the cartridge geometry by using a known activity of 131I solution spiked uniformly into a brand-new cartridge with the depth of 2 mm from the front surface. 1.5. Inhalation intake and dose calculation There were 12 radioiodine production terms in 2015 at Nuclear Research Institute of Viet Nam, once a month. The annual personal intake due to inhalation was calculated from short-term intake estimated by Eq. (1) (Klepeis 2006; Ott 1982). K
Ii
I k 1
K
ik
R
J
C
jk
tijk
(1)
k 1 j 1
where, Ii is the annual intake of subject i; Iik is the intake of subject i on the kth work day; R = 1.2 m3.h-1 is the breathing rate (IAEA 1999); Cjk is the average airborne radioactivity concentration in ME j calculated for the kth work day; tijk is the sum of the time person i spent in ME j on the kth work day; K is the total number of work day and J is the number of ME. Inhalation dose calculated from the intake multiplied by the dose coefficient for 131I vapor. For this monitoring, the value of dose coefficient for 131I vapor equals 2.0E-08 Sv.Bq-1 (IAEA 1999). 3. Results and discussion The profiles shown in Fig. 4 are an example from the dataset collected by the motion detectors on July 18th 2015 illustrating how time-ME distribute in a work day for each individual exposure. Fig. 4. Time-ME pattern of eight workers spent on July 18th 2015. The data gaps (blank between color bars) show the time workers not being in the interest microenvironments. There are eight workers, coded from W1 to W8, participated the radioiodine production in July 18th 2015. Each individual of this group has a particular work, but it can be divided into three main tasks including distilling, dividing or packing product and supervising. Their work time usually starts at 8 AM till completing the plan in the late night. For example, on July 18th 2015, it was performed from 8:00 to 22:30. This group spent most of the work time in three indoor microenvironments, room 1, room 2 and room 3 (Fig. 2). The sum of time in each room occupied by workers vary significantly. For instance, W1’s duty is to manipulate the distilling line. He mostly spent his time in room 1, 393 min compared to 19 min in room 2 and 34 min in room 3. W2 is a manager, he entered the production area for checking the workers’ task and kept the operating system works properly. Therefore, his time pattern is quite different from the others, he spent 18 min, 24 min and 15 min in room 1, room 2 and room 3 respectively during July 18 th. The worker W5, W6, and W7 sometimes entered room 3 in the afternoon in the task of handling the packed capsules and iodine solution. All the worker from W3 to W8 have the same responsibility during the work day, their main task is dividing iodine activity and packing products. Overall, time spent by the workers in room 3 generally in the early morning, in room 2 in the afternoon and in room 1 for W1. Time-ME pattern shown in this article is being the 1-min resolution. This is a quite good resolution in an exposure assessment. The time resolution of 1 min is present on Fig. 4 while the time dataset recorded by the motion detection has the 1-second resolution. Conveniently, in order to minimize the hand processing time, as we transferred the time dataset of smartphone’s file in *.srt format to Excel file, each time segment in the pattern is rounded into 1 min. As a result, it may lead to increase the total uncertainty, but its relative uncertainty quite small compared to the concentration standard deviation in an inhalation dose assessment. Therefore, we have ignored this uncertainty component in this paper. The good time-ME resolution allows the short-term dose assessment in a high accuracy. In this work, however, we do not intend to estimate individual short-term dose due to lack of continuous air concentration data. Table 2 and Fig. 5 show the sum of time spent by the group of eight workers at three microenvironments collected from smartphones in 2015. There were 12 radioiodine production terms in 2015 at Nuclear Research Institute of Viet Nam, once a month. Each term lasts for about 14 hours in a day, starting at 8:00 AM. We target to monitor eight radiation workers involved in processing 131I production. On 14 November, time data was not collected due to the technical error. Table 2. Sum of time (in minute) spent by the group of eight workers at each microenvironment collected from smartphone in 2015
10 Jan 01 Feb 14 Mar 11 Apr 23 May 20 Jun 18 Jul 22 Aug 19 Sep 17 Oct 14 Nov 12 Dec
R1 461 196 285 499 413 339 393 334 546 305 236
10 Jan 01 Feb 14 Mar 11 Apr 23 May 20 Jun 18 Jul 22 Aug 19 Sep 17 Oct 14 Nov 12 Dec
R1 82 40 54 151 77 9 74 28 21 51 26
329 30 W5 R2 23 0 3 157 7 2 65 21 3 3 7
Sum Mean
613 56
291 26
Sum Mean
4007 364
W1 R2 45 31 28 39 14 61 19 13 13 37 29
R3 41 32 58 32 38 48 34 29 43 40 -
R1 76 50 52 95 52 14 18 33 71 35 NA
395 40
496 45
R3 0 15 0 0 1 0 16 3 0 8 -
R1 181 66 126 166 159 176 201 127 350 269 107
43 4
W2 R2 8 6 38 26 12 0 24 8 0 202 NA 324 29 W6 R2 235 115 60 127 93 153 96 21 63 90 29
R3 48 6 23 27 34 11 15 22 30 0 NA
R1 77 18 40 99 73 108 65 95 66 77 107
216 20
825 75
R3 0 28 0 5 9 3 30 0 37 11 -
R1 159 45 95 178 210 161 99 81 210 128 122
W3 R2 330 104 229 466 248 209 419 130 314 249 228 2926 266 W7 R2 319 162 150 432 170 141 221 115 152 140 141
R3 52 15 52 31 31 43 19 42 13 31 -
R1 75 36 30 129 126 47 118 55 23 15 3
329 33
657 60
R3 0 7 0 26 32 7 3 18 15 3 -
R1 NA 1 4 24 6 0 10 NA NA NA NA NA
W4 R2 379 108 229 339 151 303 337 308 493 346 257 3250 295 W8 R2 NA 83 38 303 148 142 159 NA NA NA NA NA
1928 1082 123 1488 2143 111 45 873 175 98 12 135 195 11 6 125 Note: “-” indicates missing sample, NA means No Attendance
R3 0 3 0 4 4 4 28 2 2 0 47 5 R3 NA 0 0 0 0 0 0 NA NA NA NA NA 0 0
Time spent by the group in three rooms varies from person to person. These differences originate from differences between their behavior and the corresponding tasks. The mean time of W1 is over 6 hours a work day and higher in comparison with all other workers (Table 2). Moreover, we can obtain some other details about the patterns from Fig. 5 and Table 2. For example, two workers W1 and W6 primarily worked in room 1 while W3 and W4 spent most of their time in room 2. W8 had worked for 6 months, not occupied room 3 in all the shifts. This detailed data is useful in investigating worker’s behavior and allows to calculate individual short-term exposure. Also, it is probably helpful in constraining dose. Fig. 5. Statistic result shows time spent by eight radiation workers in each ME on the 131I processing days during 2015 The advantage of using the smartphone as a motion detector to monitor people is obtaining detail and real-time timeMEs for every person with the low-cost device. This task can be done with the assistance of GPS, diaries or questionnaire, but impossible in the case of indoor monitoring. The disadvantages of this method, however, is in the step of visual identifying the object’s name from captured videos and translating from smartphone’s data to computer’s processing software. Grabbed air sampling practice is performed parallel to the time-ME collecting in all ME. Detailed information of indoor air 131I concentration on operating days of 2015 at the interesting rooms is present in Table 3. In general, the 131 I vapor concentration in room 1 is higher compared to the other rooms because the distillation boxes are plotted here. The 131I vapor disperses into the indoor air probably from the distilling process.
An abnormal level of 131I concentration found on 22 August, this highest number has reached to 17290.4 Bq.m-3 on average. The considerable problem is that the similar result also happened at the two remaining rooms. This means that the concentration in these connected rooms is strongly related to each other. Table 3. Descriptive statistic of 131I concentration in three rooms at DNRI in 2015 (Bq/m3) Date
10 Jan
Room 1
Room 2
Room 3
N
Mean
SD
N
Mean
SD
N
Mean
SD
8
348.5
175.4
7
268.3
121.8
0
-
-
01 Feb
6
989.7
303.2
5
620.9
238.0
0
-
-
14 Mar
10
2580.0
1011.7
8
469.5
233.4
5
576.5
269.9
11 Apr
9
980.6
200.0
7
466.2
213.3
4
509.3
228.8
23 May
8
493.6
344.4
8
788.5
563.6
5
396.2
318.8
20 Jun
11
1027.6
682.5
11
1986.5
1754.7
4
708.2
116.4
18 Jul
9
567.5
350.6
7
212.2
103.2
4
726.4
169.6
22 Aug
7
17290.4
18741.7
8
3639.9
3197.2
4
2985.2
4026.0
19 Sep
8
955.0
876.0
8
303.3
130.1
4
459.5
277.1
17 Oct
7
2240.9
1606.0
7
1605.0
1719.0
5
364.0
328.0
14 Nov
0
-
0
-
-
0
-
-
12 Dec
9
0
-
-
1131.5
739.2
8
746.7
355.7
Based on the detailed time data in Table 2 which is collected from the smartphones, the sum of time spent at the monitored areas by the workers in 2015 calculated and showed in Table 4. Consequently, the corresponding individual 131I inhalation intake is also showed in Table 4. Where, the time spent in the MEs, in minute, is the sum of time each targeted individual exposed in the three rooms of interest; the individual intake is calculated from Eq. (1). Inhalation dose calculated from the intake multiplied by the breathing rate and the dose coefficient for 131I vapor. For this monitoring, the value of breathing rate equals 1.2 m3.h-1 and the dose coefficient for 131I vapor equals 2.0E-08 Sv.Bq-1 (IAEA 1999). Table 4. The sum of time spent in the three MEs and 131I inhalation intake of eight workers in 2015 Time spent in the MEs (min) Inhalation Intake (x103 Bq) (a)
Inhalation dose (mSv)
W1
W2
W3
W4
W5
W6
W7
W8
4731
1036
4080
3954
947
3133
3742
918
237.32
32.26
112.74
108.86
30.56
120.79
106.46
13.73
4.44
0.70
2.18
2.06
0.57
2.26
2.04
0.27(b)
(a)-calculated from the inhalation dose in 11 months multiplied with the factor of 12/11 due to the sample of one month missed, (b)- W8 worked from February to July 2015, this value indicates his inhalation dose for 6 months.
The intake differs significantly from person to person originnates from the difference between their time-ME. One of the monitored workers (worker W8) received the least intake among the group due to he has not worked from August. All the worker have exposed with an internal dose below 6 mSv, classified as low risk, and shall be monitored allowing the quantification of their annual exposures (Henrichs 2005). Especially, the worker W1 was committed an inhalation dose of 4.44 mSv, the highest compared with the others. This can be explained by looking at his time-ME pattern showed in Fig. 4, most of his time spent in room 1 where 131I concentration usually higher than the other rooms. From this result, we can confirm that the time-ME plays an important role on the estimate of inhalation dose. Based on suggested criteria for individual monitoring (IAEA 1999), a number of factors need to be taken into account in the case of committed effective doses of 1 mSv or greater in a year. That includes (i) the physical and chemical properties of the material being handled, (ii) the experience of the operation being performed and the form of the material; and (iii) the use of permanent laboratory protective equipment. Still, this study does not intend to investigate these factors. In order to obtain a better assessment and improve the method, an independent monitoring needs to be performed simultaneously for comparison. 4. Conclusions
Studying the human exposure is a laborious work. In this paper, by using an application installed on smartphones, we collected the time–ME data spent by a radiation group during 2015. This also used the portable air sampler which was combined with time data to estimate the individual indoor exposure due to inhalation of 131I vapor for eight workers working at DNRI. Based on the results and experience, we concluded that using a smartphone as a motion detector in collecting time-ME is a feasible and reliable way instead of the classic questionnaires, diary or evenly GPS-based method. Therefore, the technique presented in this paper may be an appropriate choice for integrated data collection to improve the accuracy of the data analysis. It is, however, only suitable for monitoring on fixed indoor environments and limited targeted people. Acknowledgements We would like to thank the staffs working at the Radioisotope Production Center and the Radiation Protection Center of Da Lat Nuclear Research Institute for their support to this work. References Beekhuizen, J., H. Kromhout, A. Huss, and R. Vermeulen. 2013. Performance of GPS-devices for environmental exposure assessment. Journal of Exposure Science and Environmental Epidemiology 23 (5):498-505. Bitar, A., M. Maghrabi, and A. W. Doubal. 2013. Assessment of intake and internal dose from iodine-131 for exposed workers handling radiopharmaceutical products. Applied Radiation and Isotopes 82:370-375. Broich, A., L. Gerharz, and O. Klemm. 2012. Personal monitoring of exposure to particulate matter with a high temporal resolution. Environmental Science and Pollution Research 19 (7):2959-2972. Carneiro, L. G., E. A. de Lucena, C. da Silva Sampaio, A. L. A. Dantas, W. O. Sousa, M. S. Santos, and B. M. Dantas. 2015. Internal dosimetry of nuclear medicine workers through the analysis of 131 I in aerosols. Applied Radiation and Isotopes 100:70-74. Gerharz, L. E., A. Krüger, and O. Klemm. 2009. Applying indoor and outdoor modeling techniques to estimate individual exposure to PM2. 5 from personal GPS profiles and diaries: a pilot study. Science of The Total Environment 407 (18):5184-5193. Glasgow, M. L., C. B. Rudra, E. H. Yoo, M. Demirbas, J. Merriman, P. Nayak, C. Crabtree-Ide, A. A. Szpiro, A. Rudra, J. Wactawski-Wende, and L. Mu. 2014. Using smartphones to collect time-activity data for long-term personal-level air pollution exposure assessment. J Expo Sci Environ Epidemiol (2014). Goldin, L., L. Ansher, A. Berlin, J. Cheng, D. Kanopkin, A. Khazan, M. Kisivuli, M. Lortie, E. Bunker Peterson, L. Pohl, S. Porter, V. Zeng, T. Skogstrom, M. Fragala, T. Myatt, J. Stewart, and J. Allen. 2014. Indoor Air Quality Survey of Nail Salons in Boston. Journal of Immigrant and Minority Health 16 (3):508-514. Henrichs, K. 2005. The Forthcoming ISO-Standard for the Monitoring of Workers. HEIR 2004:254. International Atomic Energy Agency (IAEA). 1999. Assessment of Occupational Exposure Due to Intakes of Radionuclides, Safety Standards Series No. RS-G-1.2. Vienna, Austria: IAEA Safety Guide, No. RS-G-1.2, Vienna, Austria. International Atomic Energy Agency (IAEA). 2000. Indirect methods for assessing intakes of radionuclides causing occupational exposure. Vienna, Austria: IAEA Safety Reports Series No. 18, Vienna, Austria. Klepeis, N. E. 2006. Modeling human exposure to air pollution. Human exposure analysis:445-470. Krajewska, G., and K. A. Pachocki. 2013. Assessment of exposure of workers to ionizing radiation from radioiodine and technetium in nuclear medicine departmental facilities. Medycyna pracy 64 (5):625-630. Kravchik, T., S. Levinson, S. Oved, S. Tsroya, O. Pelled, M. Haim, and U. German. 2008. Determination of radioiodine activity in charcoal cassettes. Applied Radiation and Isotopes 66 (6–7):972-975. Montgomery, D. 1990. Calibrating germanium detectors for assaying radioiodine in charcoal cartridges. Radioact. Radiochem. Counting Room 1 (2):47-51. Nethery, E., G. Mallach, D. Rainham, M. S. Goldberg, and A. J. Wheeler. 2014. Using Global Positioning Systems (GPS) and temperature data to generate time-activity classifications for estimating personal exposure in air monitoring studies: an automated method. Environmental Health 13 (1):33. Ott, W. R. 1982. Concepts of human exposure to air pollution. Environment International 7 (3):179-196. Steinle, S., S. Reis, and C. E. Sabel. 2013. Quantifying human exposure to air pollution—Moving from static monitoring to spatio-temporally resolved personal exposure assessment. Science of The Total Environment 443:184-193. Ton Nam, S. 2015. MotionRecorder Quickstart Guide 2013 [cited 02 January 2015]. Available from http://tonnamsoftware.com/mrec/quickstart.html. Vidal, M. V. S., A. L. A. Dantas, and B. Dantas. 2007. A methodology for auto-monitoring of internal contamination by 131I in nuclear medicine workers. Radiation protection dosimetry 125 (1-4):483-487.
Highlights
We constructed the time–microenvironment patterns with 1-minute resolution by using a smartphone application. Exposure to 131I at the dry distillation areas may lead to an acute inhalation dose significantly. Using smartphone as a motion detector in indoor exposure monitoring is a reliable method.
FIG.1
a)
b)
c)
FIG.2
a) FIG.3
b)
Indoor Inhalation Dose Monitoring Diary (daily)
Motion Detector (1 sec)
Grab air sampling (2 hr each ME)
Support
Translate Captured motion + Timestamp to Time-ME pattern
Inhalation Dose
Daily average concentration
PII: DOI: Reference:
S0969-8043(16)30293-7 http://dx.doi.org/10.1016/j.apradiso.2016.06.024 ARI7529
To appear in: Applied Radiation and Isotopes Received date: 13 January 2016 Revised date: 10 June 2016 Accepted date: 20 June 2016 Cite this article as: Tran Xuan Hoi, Huynh Truc Phuong and Nguyen Van Hung, Using Smartphone as a Motion Detector to Collect Time-microenvironment Data for estimating the inhalation dose, Applied Radiation and Isotopes, http://dx.doi.org/10.1016/j.apradiso.2016.06.024 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.
Using Smartphone as a Motion Detector to Collect Time-microenvironment Data for estimating the inhalation dose Tran Xuan Hoi1,2,*, Huynh Truc Phuong2, Nguyen Van Hung3 1
Faculty of Natural Science, Phu Yen University, 18 Tran Phu―Tuy Hoa, Viet Nam.
2
Faculty of Physics and Engineering Physics, VNUHCM–University of Science, 227 Nguyen Van Cu―Ho Chi Minh City, Viet Nam. 3
Training Center, Nuclear Research Institute, 01 Nguyen Tu Luc―Da Lat, Viet Nam.
[email protected] [email protected] [email protected] *
Corresponding author.
ABSTRACT During the production of iodine-131 from neutron irradiated tellurium dioxide by the dry distillation, a considerable amount of 131 I vapor is dispersed to the indoor air. People who routinely work at the production area may result in a significant risk of exposure to chronic intake by inhaled 131I. This study aims to estimate the inhalation dose for individuals manipulating the 131I at a radioisotope production. By using an application installed on smartphones, we collected the time–microenvironment data spent by a radiation group during work days in 2015. Simultaneously, we used a portable air sampler combined with radioiodine cartridges for grabbing the indoor air samples and then the daily averaged 131I concentration was calculated. Finally, the timemicroenvironment data jointed with the concentration to estimate the inhalation dose for the workers. The result showed that most of the workers had the annual internal dose in 16 mSv. We concluded that using smartphone as a motion detector is a possible and reliable way instead of the questionnaires, diary or GPS-based method. It is, however, only suitable for monitoring on fixed indoor environments and limited the targeted people. Keywords: Iodine-131; Motion detector; time-microenvironment; inhalation dose
1. Introduction Radioiodine-131 is most commonly used for diagnostics and therapy in medicine. People who are occupationally exposed to an internal uptake of iodine includes medical staffs working at nuclear medicine departments and radiation workers of radioisotopes productions. Routine handling of solutions containing radioiodine may result in a significant risk of exposure of the subjects to chronic intake by inhalation of aerosols (IAEA 1999; Bitar et al. 2013; Carneiro et al. 2015; Vidal et al. 2007; Krajewska and Pachocki 2013). A published paper showed that the annual internal effective dose for some workers were above 1 mSv and a worker reached 7.66 mSv with high-risk classification; and this worker must be monitored individually (Bitar et al. 2013). One of the situations needed to be monitored for internal exposure is handling of large quantities of radiopharmaceuticals, such as 131I for therapy (IAEA 1999). The choice of monitoring method depends on factors such as the availability of instrumentation, the analyzes costs, as well as on the sensitivity that is needed. Two in vitro methods were usually used aiming to assess the internal doses due to inhalation intake of 131I: first way was based on the measurement of 131I activity in 24-hour urine samples, and the second one was based on workplace monitoring the 131I aerosol concentration of the indoor air (IAEA 1999). To model exposure to airborne elements, one uses the conceptually simple approach of matching the locations that each exposed person visits with the time-averaged or dynamic air pollutant concentrations that are thought to exist in each visited location (Klepeis 2006; Ott 1982). For obtaining the time people spent in indoor visited locations, researchers used some approaches such as Questionnaire, diaries or GPS-based method (Goldin et al. 2014; Steinle et al. 2013; Carneiro et al. 2015; Glasgow et al. 2014). Diary was usually used as a conventional tool for construct the time-activity pattern. A limiting factor, however, lay in the diaries that only provided restricted information about indoor environments and personal behavior, so the model produces uncertain values in consequence. The complexity and accuracy of the indoor model are limited by the diary accuracy (Gerharz et al. 2009). Using GPS-based method may give an undesirable location precision. GPS-generated time-activity has been well tested and provides an average precision of 7 m in typical urban conditions (Nethery et al. 2014). The precision of 7 m may not be suitable for individual exposure monitoring in normal dimension rooms. In addition, the dependence of GPS-accuracy on many factors such as unfavorable or clustered satellite positions, or ionospheric disturbances in a particular local area is the obstacle must be considered. Another major cause of error is obstruction of the satellite signal. Therefore, GPS-errors will place a participant at the wrong location (Beekhuizen et al. 2013). All the models mentioned above are not including the exact geographical position of the individuals. Researchers who used these methods have to minimize the inconvenience to routine work as well as would get the low accuracy of time-microenvironment (time-ME) data in case of indoor environment monitoring. To avoid these disadvantages, some authors advised that using mobile phone with the help of its application software in tracking environment. Smartphone with inbuilt camera can be used for monitoring fixed places and could lead to more detailed dataset (Broich et al. 2012; Gerharz et al. 2009). Alongside, static air samplers may be used to determine the concentration of airborne radioactive material, which can be combined with site specific assumptions about the physicochemical form of the material and the breathing rate and exposure time of the worker to estimate inhalation intakes (IAEA 1999). In this research, we set the Symbian-based smartphone as a motion detector by installing a commercial application named MotionRecorder which is developed by Ton Nam Software (Ton Nam 2013). This monitoring system uses the built-in camera to detect movements in the surrounding area using an advanced motion detection algorithm without activating the built-in GPS. The application also indicates real-time on the phones display where the movement is detected. The aim of this paper is using low-cost and unwearable devices including a mobile phone’s motion detection application and portable air sampler for estimating individual inhalation dose of workers processing 131I production. 2. Materials and Methods The inhalation dose of workers during the 131I processing period was estimated by using the methodology described in Fig. 1. Fig. 1. Diagram of the methodology. Diary is added when missing data from the Motion Detector, the air sample was grabbed in 2 hr intervals in a particular ME. Daily Individual internal dose was calculated from the daily average of 131I concentration in each ME jointed with time-ME. 1.1. The studied MEs
The study was carried out at the radioiodine production area of Da Lat Nuclear Research Institute in Viet Nam which consists of three rooms (Fig. 2): 131I distilling room (room 1), activity dividing room (room 2) and hot cell room (room 3). Each of these MEs has the same dimension of 6m(L)x5m(W)x4m(H) without any window. The three rooms are connected by two internal doors with room 1 and workers can enter the rooms from an enclosed corridor. Exhaust and air supply systems run continuously during the work time. These MEs are indoor environments and not large. Therefore, in this study, it is assumed that the air concentration in each room is uniform (IAEA 1999). Fig. 2. Experimental arrangement of motion detectors located next to doors in interesting rooms in order to minimize the captured view. Two smartphones needed to monitor indoor exposure of the workers in three rooms, (a) in room 3, (b, c) in room 2. The clock on these phones was updated via Wi-Fi. 1.2. Time-ME collection In order to record time-ME spent by each targeted worker, we used a motion-sensitive application installed in Symbian-based smartphones, named MotionRecorder with light size on disk after installed (Ton Nam 2013). Table 1 shows some characteristics of the “app”. Table 1. MotionRecorder version 1.0 property Items Operating System requirement Size on disk after installed Motion detection sensitivity level available Pixel change sensitivity level available Video resolution output selection Record after movement detected Time-stamp file format Run in background Record audio/mute
Parameter Symbian Anna/Belle 76 kB 0100 0100 HD (1200720) maximum 010 sec *.srt -
Option Adjustable Adjustable Adjustable Adjustable Fixed Available Available
The application turns the phones into a motion-activated video camera. It is run based on the change of two parameters: pixel change sensitivity and motion detection sensitivity. The software allows user can adjust these parameters manually in order to record the motion as expected, and enable smartphone running in background mode to reduce energy usage. Nevertheless, to make sure the smartphone not being in the energy shortage, it was connected to a battery storage (Fig. 2) so that this system could live during a work day. As running, all the movement happening in front of the phone is recorded to a video file subtitled with time and date. When workers enter and leave the monitored rooms, the software records the motion and attaches the real time in hh:mm:ss as soon as these motions occur (Fig. 3). In this research, we used two smartphones in order to monitor the motion occurred at the four doors, each of them is responsible for observing the events at two doors next to each other. The monitor was fixed to the wall next to the doors to confirm that the monitored view was minimized (Fig. 2). The time data and captured video collected by smartphones are manually translated into Microsoft Excel ® file in order to process consequently. Fig. 3. Smartphone installed the Motion Recorder application used as a motion detector to collect time-ME data, (a) the application on standby mode and (b) data file shows timestamp attached to each captured scene. 1.3. Air sampling experimental arrangement Experimental arrangement was performed in the workplace with respect to minimizing the inconvenience to routine work. In regularly, the time of 1.5 hour needed to run a distillation batch in an oven at room 1. At room 2, the group starts to work at noon. People rarely occupied room 3 or the duration is short. A portable air sampler that is Eberline Model RAS-1 and the activated carbon cartridges impregnated with TEDA (Hi-Q Model TC-12) were used to sample the indoor air. The cartridge holder was settled at several position in the rooms near the breathing zone at 1.5m height from the floor (IAEA 2000). The sampling time was about 5 minutes per sample at the flow rate of 70 L/min. The cartridges were sampled in the duration of 510 min corresponding the volume of 0.350.70 m3. 1.4. Determination of 131I in cartridges The cartridges containing 131I were counted at DNRI by gamma spectrometry with a Canberra HPGe coaxial detector system supplied by Oxford Instruments Inc. Nuclear Measurement Group (Model No. CPVDS 30-30185) operated with the pulse height analyzer software Oxford PCA multiport 16K.
Based on some published papers (Montgomery 1990; Kravchik et al. 2008), when relatively small volumes of air are being sampled, it is assumed that most radioiodine is concentrated in the front side of the cassette, greater than 95% of the radioiodine in charcoal cartridge samples was within the front 5 mm and average depth was about 2 mm. Therefore, the detection system used in this counting was calibrated for the cartridge geometry by using a known activity of 131I solution spiked uniformly into a brand-new cartridge with the depth of 2 mm from the front surface. 1.5. Inhalation intake and dose calculation There were 12 radioiodine production terms in 2015 at Nuclear Research Institute of Viet Nam, once a month. The annual personal intake due to inhalation was calculated from short-term intake estimated by Eq. (1) (Klepeis 2006; Ott 1982). K
Ii
I k 1
K
ik
R
J
C
jk
tijk
(1)
k 1 j 1
where, Ii is the annual intake of subject i; Iik is the intake of subject i on the kth work day; R = 1.2 m3.h-1 is the breathing rate (IAEA 1999); Cjk is the average airborne radioactivity concentration in ME j calculated for the kth work day; tijk is the sum of the time person i spent in ME j on the kth work day; K is the total number of work day and J is the number of ME. Inhalation dose calculated from the intake multiplied by the dose coefficient for 131I vapor. For this monitoring, the value of dose coefficient for 131I vapor equals 2.0E-08 Sv.Bq-1 (IAEA 1999). 3. Results and discussion The profiles shown in Fig. 4 are an example from the dataset collected by the motion detectors on July 18th 2015 illustrating how time-ME distribute in a work day for each individual exposure. Fig. 4. Time-ME pattern of eight workers spent on July 18th 2015. The data gaps (blank between color bars) show the time workers not being in the interest microenvironments. There are eight workers, coded from W1 to W8, participated the radioiodine production in July 18th 2015. Each individual of this group has a particular work, but it can be divided into three main tasks including distilling, dividing or packing product and supervising. Their work time usually starts at 8 AM till completing the plan in the late night. For example, on July 18th 2015, it was performed from 8:00 to 22:30. This group spent most of the work time in three indoor microenvironments, room 1, room 2 and room 3 (Fig. 2). The sum of time in each room occupied by workers vary significantly. For instance, W1’s duty is to manipulate the distilling line. He mostly spent his time in room 1, 393 min compared to 19 min in room 2 and 34 min in room 3. W2 is a manager, he entered the production area for checking the workers’ task and kept the operating system works properly. Therefore, his time pattern is quite different from the others, he spent 18 min, 24 min and 15 min in room 1, room 2 and room 3 respectively during July 18 th. The worker W5, W6, and W7 sometimes entered room 3 in the afternoon in the task of handling the packed capsules and iodine solution. All the worker from W3 to W8 have the same responsibility during the work day, their main task is dividing iodine activity and packing products. Overall, time spent by the workers in room 3 generally in the early morning, in room 2 in the afternoon and in room 1 for W1. Time-ME pattern shown in this article is being the 1-min resolution. This is a quite good resolution in an exposure assessment. The time resolution of 1 min is present on Fig. 4 while the time dataset recorded by the motion detection has the 1-second resolution. Conveniently, in order to minimize the hand processing time, as we transferred the time dataset of smartphone’s file in *.srt format to Excel file, each time segment in the pattern is rounded into 1 min. As a result, it may lead to increase the total uncertainty, but its relative uncertainty quite small compared to the concentration standard deviation in an inhalation dose assessment. Therefore, we have ignored this uncertainty component in this paper. The good time-ME resolution allows the short-term dose assessment in a high accuracy. In this work, however, we do not intend to estimate individual short-term dose due to lack of continuous air concentration data. Table 2 and Fig. 5 show the sum of time spent by the group of eight workers at three microenvironments collected from smartphones in 2015. There were 12 radioiodine production terms in 2015 at Nuclear Research Institute of Viet Nam, once a month. Each term lasts for about 14 hours in a day, starting at 8:00 AM. We target to monitor eight radiation workers involved in processing 131I production. On 14 November, time data was not collected due to the technical error. Table 2. Sum of time (in minute) spent by the group of eight workers at each microenvironment collected from smartphone in 2015
10 Jan 01 Feb 14 Mar 11 Apr 23 May 20 Jun 18 Jul 22 Aug 19 Sep 17 Oct 14 Nov 12 Dec
R1 461 196 285 499 413 339 393 334 546 305 236
10 Jan 01 Feb 14 Mar 11 Apr 23 May 20 Jun 18 Jul 22 Aug 19 Sep 17 Oct 14 Nov 12 Dec
R1 82 40 54 151 77 9 74 28 21 51 26
329 30 W5 R2 23 0 3 157 7 2 65 21 3 3 7
Sum Mean
613 56
291 26
Sum Mean
4007 364
W1 R2 45 31 28 39 14 61 19 13 13 37 29
R3 41 32 58 32 38 48 34 29 43 40 -
R1 76 50 52 95 52 14 18 33 71 35 NA
395 40
496 45
R3 0 15 0 0 1 0 16 3 0 8 -
R1 181 66 126 166 159 176 201 127 350 269 107
43 4
W2 R2 8 6 38 26 12 0 24 8 0 202 NA 324 29 W6 R2 235 115 60 127 93 153 96 21 63 90 29
R3 48 6 23 27 34 11 15 22 30 0 NA
R1 77 18 40 99 73 108 65 95 66 77 107
216 20
825 75
R3 0 28 0 5 9 3 30 0 37 11 -
R1 159 45 95 178 210 161 99 81 210 128 122
W3 R2 330 104 229 466 248 209 419 130 314 249 228 2926 266 W7 R2 319 162 150 432 170 141 221 115 152 140 141
R3 52 15 52 31 31 43 19 42 13 31 -
R1 75 36 30 129 126 47 118 55 23 15 3
329 33
657 60
R3 0 7 0 26 32 7 3 18 15 3 -
R1 NA 1 4 24 6 0 10 NA NA NA NA NA
W4 R2 379 108 229 339 151 303 337 308 493 346 257 3250 295 W8 R2 NA 83 38 303 148 142 159 NA NA NA NA NA
1928 1082 123 1488 2143 111 45 873 175 98 12 135 195 11 6 125 Note: “-” indicates missing sample, NA means No Attendance
R3 0 3 0 4 4 4 28 2 2 0 47 5 R3 NA 0 0 0 0 0 0 NA NA NA NA NA 0 0
Time spent by the group in three rooms varies from person to person. These differences originate from differences between their behavior and the corresponding tasks. The mean time of W1 is over 6 hours a work day and higher in comparison with all other workers (Table 2). Moreover, we can obtain some other details about the patterns from Fig. 5 and Table 2. For example, two workers W1 and W6 primarily worked in room 1 while W3 and W4 spent most of their time in room 2. W8 had worked for 6 months, not occupied room 3 in all the shifts. This detailed data is useful in investigating worker’s behavior and allows to calculate individual short-term exposure. Also, it is probably helpful in constraining dose. Fig. 5. Statistic result shows time spent by eight radiation workers in each ME on the 131I processing days during 2015 The advantage of using the smartphone as a motion detector to monitor people is obtaining detail and real-time timeMEs for every person with the low-cost device. This task can be done with the assistance of GPS, diaries or questionnaire, but impossible in the case of indoor monitoring. The disadvantages of this method, however, is in the step of visual identifying the object’s name from captured videos and translating from smartphone’s data to computer’s processing software. Grabbed air sampling practice is performed parallel to the time-ME collecting in all ME. Detailed information of indoor air 131I concentration on operating days of 2015 at the interesting rooms is present in Table 3. In general, the 131 I vapor concentration in room 1 is higher compared to the other rooms because the distillation boxes are plotted here. The 131I vapor disperses into the indoor air probably from the distilling process.
An abnormal level of 131I concentration found on 22 August, this highest number has reached to 17290.4 Bq.m-3 on average. The considerable problem is that the similar result also happened at the two remaining rooms. This means that the concentration in these connected rooms is strongly related to each other. Table 3. Descriptive statistic of 131I concentration in three rooms at DNRI in 2015 (Bq/m3) Date
10 Jan
Room 1
Room 2
Room 3
N
Mean
SD
N
Mean
SD
N
Mean
SD
8
348.5
175.4
7
268.3
121.8
0
-
-
01 Feb
6
989.7
303.2
5
620.9
238.0
0
-
-
14 Mar
10
2580.0
1011.7
8
469.5
233.4
5
576.5
269.9
11 Apr
9
980.6
200.0
7
466.2
213.3
4
509.3
228.8
23 May
8
493.6
344.4
8
788.5
563.6
5
396.2
318.8
20 Jun
11
1027.6
682.5
11
1986.5
1754.7
4
708.2
116.4
18 Jul
9
567.5
350.6
7
212.2
103.2
4
726.4
169.6
22 Aug
7
17290.4
18741.7
8
3639.9
3197.2
4
2985.2
4026.0
19 Sep
8
955.0
876.0
8
303.3
130.1
4
459.5
277.1
17 Oct
7
2240.9
1606.0
7
1605.0
1719.0
5
364.0
328.0
14 Nov
0
-
0
-
-
0
-
-
12 Dec
9
0
-
-
1131.5
739.2
8
746.7
355.7
Based on the detailed time data in Table 2 which is collected from the smartphones, the sum of time spent at the monitored areas by the workers in 2015 calculated and showed in Table 4. Consequently, the corresponding individual 131I inhalation intake is also showed in Table 4. Where, the time spent in the MEs, in minute, is the sum of time each targeted individual exposed in the three rooms of interest; the individual intake is calculated from Eq. (1). Inhalation dose calculated from the intake multiplied by the breathing rate and the dose coefficient for 131I vapor. For this monitoring, the value of breathing rate equals 1.2 m3.h-1 and the dose coefficient for 131I vapor equals 2.0E-08 Sv.Bq-1 (IAEA 1999). Table 4. The sum of time spent in the three MEs and 131I inhalation intake of eight workers in 2015 Time spent in the MEs (min) Inhalation Intake (x103 Bq) (a)
Inhalation dose (mSv)
W1
W2
W3
W4
W5
W6
W7
W8
4731
1036
4080
3954
947
3133
3742
918
237.32
32.26
112.74
108.86
30.56
120.79
106.46
13.73
4.44
0.70
2.18
2.06
0.57
2.26
2.04
0.27(b)
(a)-calculated from the inhalation dose in 11 months multiplied with the factor of 12/11 due to the sample of one month missed, (b)- W8 worked from February to July 2015, this value indicates his inhalation dose for 6 months.
The intake differs significantly from person to person originnates from the difference between their time-ME. One of the monitored workers (worker W8) received the least intake among the group due to he has not worked from August. All the worker have exposed with an internal dose below 6 mSv, classified as low risk, and shall be monitored allowing the quantification of their annual exposures (Henrichs 2005). Especially, the worker W1 was committed an inhalation dose of 4.44 mSv, the highest compared with the others. This can be explained by looking at his time-ME pattern showed in Fig. 4, most of his time spent in room 1 where 131I concentration usually higher than the other rooms. From this result, we can confirm that the time-ME plays an important role on the estimate of inhalation dose. Based on suggested criteria for individual monitoring (IAEA 1999), a number of factors need to be taken into account in the case of committed effective doses of 1 mSv or greater in a year. That includes (i) the physical and chemical properties of the material being handled, (ii) the experience of the operation being performed and the form of the material; and (iii) the use of permanent laboratory protective equipment. Still, this study does not intend to investigate these factors. In order to obtain a better assessment and improve the method, an independent monitoring needs to be performed simultaneously for comparison. 4. Conclusions
Studying the human exposure is a laborious work. In this paper, by using an application installed on smartphones, we collected the time–ME data spent by a radiation group during 2015. This also used the portable air sampler which was combined with time data to estimate the individual indoor exposure due to inhalation of 131I vapor for eight workers working at DNRI. Based on the results and experience, we concluded that using a smartphone as a motion detector in collecting time-ME is a feasible and reliable way instead of the classic questionnaires, diary or evenly GPS-based method. Therefore, the technique presented in this paper may be an appropriate choice for integrated data collection to improve the accuracy of the data analysis. It is, however, only suitable for monitoring on fixed indoor environments and limited targeted people. Acknowledgements We would like to thank the staffs working at the Radioisotope Production Center and the Radiation Protection Center of Da Lat Nuclear Research Institute for their support to this work. References Beekhuizen, J., H. Kromhout, A. Huss, and R. Vermeulen. 2013. Performance of GPS-devices for environmental exposure assessment. Journal of Exposure Science and Environmental Epidemiology 23 (5):498-505. Bitar, A., M. Maghrabi, and A. W. Doubal. 2013. Assessment of intake and internal dose from iodine-131 for exposed workers handling radiopharmaceutical products. Applied Radiation and Isotopes 82:370-375. Broich, A., L. Gerharz, and O. Klemm. 2012. Personal monitoring of exposure to particulate matter with a high temporal resolution. Environmental Science and Pollution Research 19 (7):2959-2972. Carneiro, L. G., E. A. de Lucena, C. da Silva Sampaio, A. L. A. Dantas, W. O. Sousa, M. S. Santos, and B. M. Dantas. 2015. Internal dosimetry of nuclear medicine workers through the analysis of 131 I in aerosols. Applied Radiation and Isotopes 100:70-74. Gerharz, L. E., A. Krüger, and O. Klemm. 2009. Applying indoor and outdoor modeling techniques to estimate individual exposure to PM2. 5 from personal GPS profiles and diaries: a pilot study. Science of The Total Environment 407 (18):5184-5193. Glasgow, M. L., C. B. Rudra, E. H. Yoo, M. Demirbas, J. Merriman, P. Nayak, C. Crabtree-Ide, A. A. Szpiro, A. Rudra, J. Wactawski-Wende, and L. Mu. 2014. Using smartphones to collect time-activity data for long-term personal-level air pollution exposure assessment. J Expo Sci Environ Epidemiol (2014). Goldin, L., L. Ansher, A. Berlin, J. Cheng, D. Kanopkin, A. Khazan, M. Kisivuli, M. Lortie, E. Bunker Peterson, L. Pohl, S. Porter, V. Zeng, T. Skogstrom, M. Fragala, T. Myatt, J. Stewart, and J. Allen. 2014. Indoor Air Quality Survey of Nail Salons in Boston. Journal of Immigrant and Minority Health 16 (3):508-514. Henrichs, K. 2005. The Forthcoming ISO-Standard for the Monitoring of Workers. HEIR 2004:254. International Atomic Energy Agency (IAEA). 1999. Assessment of Occupational Exposure Due to Intakes of Radionuclides, Safety Standards Series No. RS-G-1.2. Vienna, Austria: IAEA Safety Guide, No. RS-G-1.2, Vienna, Austria. International Atomic Energy Agency (IAEA). 2000. Indirect methods for assessing intakes of radionuclides causing occupational exposure. Vienna, Austria: IAEA Safety Reports Series No. 18, Vienna, Austria. Klepeis, N. E. 2006. Modeling human exposure to air pollution. Human exposure analysis:445-470. Krajewska, G., and K. A. Pachocki. 2013. Assessment of exposure of workers to ionizing radiation from radioiodine and technetium in nuclear medicine departmental facilities. Medycyna pracy 64 (5):625-630. Kravchik, T., S. Levinson, S. Oved, S. Tsroya, O. Pelled, M. Haim, and U. German. 2008. Determination of radioiodine activity in charcoal cassettes. Applied Radiation and Isotopes 66 (6–7):972-975. Montgomery, D. 1990. Calibrating germanium detectors for assaying radioiodine in charcoal cartridges. Radioact. Radiochem. Counting Room 1 (2):47-51. Nethery, E., G. Mallach, D. Rainham, M. S. Goldberg, and A. J. Wheeler. 2014. Using Global Positioning Systems (GPS) and temperature data to generate time-activity classifications for estimating personal exposure in air monitoring studies: an automated method. Environmental Health 13 (1):33. Ott, W. R. 1982. Concepts of human exposure to air pollution. Environment International 7 (3):179-196. Steinle, S., S. Reis, and C. E. Sabel. 2013. Quantifying human exposure to air pollution—Moving from static monitoring to spatio-temporally resolved personal exposure assessment. Science of The Total Environment 443:184-193. Ton Nam, S. 2015. MotionRecorder Quickstart Guide 2013 [cited 02 January 2015]. Available from http://tonnamsoftware.com/mrec/quickstart.html. Vidal, M. V. S., A. L. A. Dantas, and B. Dantas. 2007. A methodology for auto-monitoring of internal contamination by 131I in nuclear medicine workers. Radiation protection dosimetry 125 (1-4):483-487.
Highlights
We constructed the time–microenvironment patterns with 1-minute resolution by using a smartphone application. Exposure to 131I at the dry distillation areas may lead to an acute inhalation dose significantly. Using smartphone as a motion detector in indoor exposure monitoring is a reliable method.
FIG.1
a)
b)
c)
FIG.2
a) FIG.3
b)
Indoor Inhalation Dose Monitoring Diary (daily)
Motion Detector (1 sec)
Grab air sampling (2 hr each ME)
Support
Translate Captured motion + Timestamp to Time-ME pattern
Inhalation Dose
Daily average concentration