development of textile proximity sensor for medication adherence management system.
Introduction medication compliance is generally expressed by the amount and number of drugs prescribed by doctors and pharmacists .
Adherence to medication adherence is an important behavior for effective treatment of diseases and recovery of health .
However, it does not follow the compliance of drug treatment as expected by doctors and pharmacists.
This leads to problems with the reliability of doctors and pharmacists, but more importantly, it can lead to problems with the health of patients.
In a study conducted in the United States (US)
Identify the importance of drug adherence in terms of risk of death and medical expenses for patients participating in the American Medical Center program.
Therefore, the risk of death in patients with medication compliance below 80% is 3 times higher.
In addition, the number of hospital visits was 86%, and the number of emergency traffic visits increased by 50%. 3].
In order to solve the problem
Persistence, various studies have been carried out on the system that helps drug persistence.
Studies have been conducted to assess whether the wearable device is manufactured and adhered to the drug, and it includes a study using an inertial sensor to wear on the wrist to detect adhesion to the drug [4-5].
In addition, a study to predict drug adhesion using piezoelectric sensors and vibration sensors .
And a study was carried out to measure the adhesion to the ingestible sensor 7].
In addition, drug compliance studies were conducted, including the use of 3-
Shaft acceleration sensor [8-9]
And use the load cell .
On the other hand, wearable technology based on conductive texture has attracted attention in various fields such as medical 
Commercial, military and aerospace industries.
This is because, technically, it is only possible for existing electronic devices to make a limited part of the existence.
In addition, since the conduction system is human-
One advantage is that consumers can use eco-friendly textiles without causing user discomfort.
Recently, a study on medical care using conductive textiles12-17].
In particular, research on human cyclical behavior and biological activities is under way.
As a staged action, it is necessary to study sitting posture. 18-20]
Walking study [21-22]
Textile sensors are used.
Research on biometrics23-29]
Also in progress.
With the development of conductive textiles and the use of fibers, human life will be further improved in the future.
In this study, a textile sensor was manufactured by using the conductive etextile to achieve a drug adherence management system.
There were many sensors in previous studies, large and heavy, mainly the study of mechanical systems.
We propose a system of conductive textile, which is a simple, lightweight and humanized textile.
Friendly materials to solve the above problems.
The structure of this paper is as follows: Section 2 introduces the principle and implementation of the conductive fabric proximity sensor.
Section III describes the proposed system.
After Sections 4 and 5, we describe and discuss the experiments and results respectively.
Section 6 summarizes our study. 2.
Conductive textile proximity sensor 2.
1 conductive materials such as nickel coating for conductive textiles (Ni)or cooper (Cu)
On textiles, electricity is allowed to pass through textiles.
The structure of conductive textiles is shown in the figure. 1.
It has the advantages of conductive and shielding effects, is easy to manufacture sensors, and can be made in various forms.
In this study, we designed a system to evaluate drug adhesion by manufacturing textile proximity sensors using conductors.
Conductive textile model is W-290-PCN (
Ajin electronics, Busan, South Korea)
The properties of conductive textiles are shown in Table 1. 2.
Textile proximity sensors derive capacitance values based on capacitance theory. Fig. 2.
A parallel plate capacitor method for a basic capacitor is shown.
The formula for obtaining the capacitance value of parallel plate capacitors is the equation (1). C = [[[epsilon]. sub. r]x[[? ? ]. sub. 0]xA]/d(1)[[epsilon]. sub. r]
Delectric constants of materials located between two plates, and [[epsilon]. sub. 0]
Is the dielectric constant of the space (8. 85 x [10. sup. -12]F/m).
A is the area of the plate (width xlength)
, D is the distance between the two boards (in meters).
However, in the sensor implementation ,[[epsilon]. sub. r],[[epsilon]. sub. 0]
, A does not change, so the capacitance value is determined by d.
Basic capacitors of two parallel plates in figure 2
Month, fringeeffect happened.
The stripe effect affects the measurement accuracy of the system and occurs near the edge of the plate.
So in this study,
To minimize the effect of the edge effect, a type capacitor sensor has been manufactured .
The sensors manufactured consist of sensor parts, shielding parts and materials. A 5x5[cm. sup. 2]
The material is a square medical textile bandage.
The sensor part is 3x3 [cm. sup. 2]
Round size, suitable for the size of the medicine box, the shielding part is made of 5x5 [cm. sup. 2]
Round size and ring type of width 0.
As shown in figure 5 cm3. 3.
Medication compliance management system we designed the whole system based on textile proximity sensors designed for medication compliance. Fig.
4 shows the drug adherence management system.
In order to transmit the capacitance value generated by the textilesensor to the personal computer (PC)
, Custom printed circuit board (cusotmPCB)
Using 121 chip based on synchronous transmission mode (STM)
BC417 chip for Bluetooth (BTH)communication.
The custom PCB is shown in Figure 1.
5. The battery can work for up to 8 hours.
In previous research31]
, There is a problem that because the cartridge is placed in the position of the texture sensor, the data is unstable and noise is generated due to PCB and battery.
To solve this problem, use the Replicator 2 (
MakerBot, New York City, United States of America)
3D printer model.
The sensor barrier and PCB protection design are shown in Figure 1. 6.
A 5 cm wide, 5 cm long, 0 sensor barrier is designed.
The height is 2 cm depending on the sensor size.
The wall part of the sensor barrier is designed to be 3cm width, 3cm length, 1.
According to the size of the medicine box, the height is 3 cm.
The PCB guard is designed to be 4 cm wide, 3 cm long, 1.
According to the size of the PCB, the height is 3 cm.
And can connect the power cord and sensor cable through 1 hole. 5cmx0. 8cm and2cmx0.
5 cm respectively. 4. Experiments 4.
1 textile proximity sensor evaluation in previous studies we used ccr-8110G (
GW Instek, New Taipei City, Taiwan)
Confirmation results of textile sensors .
In addition, SF-using an electronic compact scale model-400C (
Su Fei electronics, Jiangsu, China)
Confirm the weight data according to the number of pills.
The performance of the electronic compact scale is shown in Table 2.
The pill used is 15T (
Tablet, 6, 000mgx15)multi-
Vitamin EnerHeim (
NutriloGmbH, Port of coques, Germany).
As a result, as shown in the figure, a graph is shown linearly.
7, the average weight of the pill is 6. 036g([+ or-]0. 064).
According to the pill data measured with the electronic scale, we use the textile proximity sensor we implemented to obtain the data according to the number of pills and correlate between these two data, to assess whether our sensors are relevant to the standards of electronic compact scales. 4.
To evaluate medication compliance, we designed a scenario for patients taking medication.
First, the patient measured the data of the sensor before taking the medicine.
Next, the patient takes the medicine pill box and puts the pill box on the textile sensor.
Finally, the sensor is processed.
Table 3 shows the steps of the medication scene. 4.
3 The Data Communication capacitance value is converted into a digital value and transmitted to the PC to display the capacitance value, which is the data generated when the cartridge is placed on the sensor.
In order to convert the digital value, the acupcb PCB passes through I (2)
C. Communication and data conversion through ADC.
The data is then transferred to the PC via bth communication.
In order to check the transmission capacitance value, the data of textile proximity sensor can be confirmed and stored in real time
Time to implement an application using C language in Visual Studio 2017, as shown in the figure8. 4.
4 Data collection based on the constructed system, we will collect the change data of the capacitance value every time we take the medicine.
We set the data sample to 100Hz to check the data in real time
Timing through Capplication, and data collection through the storage function.
The experiment carried out a total of 7 experiments, from 15T when the pill was full to 0 T when the pill was empty, according to the medication scenario. 4.
5 The Chart of the original data obtained through the experiment by signal processing is shown in the figure9. Fig. 9(a)
It is in the case of pills, that is, the state in which the pills are filled at 15 T. Fig. 9(b)
Is a state in which the pill is empty in pillcase by repeatedly taking the drug scene (Table 3). Fig. 9(c)
Showsa takes a series of medication procedures that take birth control pills and put them down.
We apply the mobile average filter (MAF, N=15)
Reduce noise and stair phenomena as the data is affected by the stair phenomenon and the noise in the collected data. And Fig. 9(c)
Detected using peak detection algorithm.
After peak detection, the time required to stabilize the data is 1 second and the data value is 1.
5 seconds after extraction in the stable area of Interst (ROI).
Take the mean from the extracted ROI data and act as an evaluation indicator for drug compliance. 4.
6 Statistical analysis to confirm the correlation between pill data for electronic compact scales and textile proximity sensors based on conductive textiles.
We performed Pearson correlation analysis on two sets of data using SPSS statistical version 18. 0 (
IBM, New York, United States)
The correlation between the two sets of data was evaluated. 5.
Results and discussion we conducted an experiment to evaluate whether there was a change in the number of drug adherence based on the drug situation.
There were a total of 7 experiments, and the results were shown in Table 4.
The data value of each experiment in Table 4 represents the capacitance value (pF)
As the number of pills decreases, the capacitance value decreases gradually.
However, although the number of pills in Table 4 has decreased, the portion with the same data value or the portion of the data increase has been confirmed.
In this question, we considered y-
The axis in the program code of the C application is set to too insensitive.
This problem will be solved by checking and correcting the problem of the code.
Data for system measurements proposed in Table 4.
In order to evaluate the stability of the whole system, the values of each experimental data were averaged.
The average capacitance value of each pill quantity is derived.
The change of the data is represented by the chart shown in the figure. 10. Fig.
10 shows that the capacitance value gradually increases as the pill increases.
We then evaluated the correlation between the proposed system and the electronic compact scale by pearson\'s correlation.
Pearson-related results of experimental data and electronic scale data are shown in Table 5.
The results show that the lowest correlation coefficient is 0.
932, the highest correlation coefficient is 0.
963 per experiment.
In addition, the correlation coefficient is 0 for the average data.
956, the results were statistically significant in all experiments .
Therefore, we conclude that our system is able to assess the compliance of the drug when taking it. 6.
Conclusion in this study, a drug adherence management system was proposed using textile proximity sensors made of conductive textures.
The results of the study show that our conclusion is that in our system, the number of pills reduced each time a drug is taken can be evaluated.
However, we have found that small problems that need to be more sensitive to the solution of the system.
Further research is that after solving the problem of the system, we will add a service function to work with smart devices in the android app to help drug adherence patients maintain drug adherence.
Also, we are trying to use our system in a real environmentlife.
Finally, we expect that the proposed system based on textile proximity sensors can be used for various studies. References 
Osterberg, Lars and Terrence Blaschke, persistent treatment, New England Journal of Medicine, Volume 1353, no. 5,487-497, 2005. Article (CrossRef Link)Cramer, J. A. , Roy, A. , Burrell, A. , Fairchild, C. J. ,Fuldeore, M. J. , Ollendorf, D. A. , & Wong, P.
K, \"medical compliance and persistence: Terms and Definitions\", health value, volume 111, no. 1, 44-47, 2008. Article (CrossRef Link)Faught, E. , Duh, M. S. , Weiner, J. R. , Guerin, A.
, And Cunnington, M.
C, \"findings of increased mortality in non-compliance with epilepsy drugs and ransom studies\", neurology, Vol. 71, no. 20, 1572-1578, 2008. Article (CrossRef Link)Chen, C.
, Kehtarnavaz, N.
, & \"Drug bottle drug adhesion monitoring system based on wearable inertial sensor\" in Jafari, R, Proc \".
Engineering major, Medical and Biological Society (EMBC)
2014 Annual IEEE International Conference, p. 4983-
4986, IEEE, 8 months. Article (CrossRefLink)Kalantarian, H. , Alshurafa, N.
, & Sarrafzadeh, M, \"use a smartwatch to detect gestures related to drug persistence --
IEEE Journal of sensors, Volume 1. 16,no. 4, 1054-1061, 2016. Article (CrossRefLink)Kalantarian, H. , Motamed, B. , Alshurafa, N.
, & Sarrafzadeh, M, \"wearable sensor system for drug adhesion prediction\", Medical Artificial Intelligence, Vol. 69, 43-52,2016. Article (CrossRefLink)
Hafezi, Hooman, et al, \"an edible sensor for measuring drug compliance\", IEEE biomedical engineering transaction, Volume 162, no. 1, 99-109, 2015. Article (CrossRefLink)Li, J. , Peplinski, S. J. , Nia, S. M.
, & \"Interoperable medicine box system for smart medical persistence\" in Farajidavar, A, Proc \".
Engineering major, Medical and Biological Society (EMBC)
2014 Annual IEEE International Conference, p. 1386-
1389, IEEE, 8 months. Article (CrossRefLink)Lee, M. L. , & Dey, A. K. , \"Real-
\"Improve time feedback on taking medicine,\" in Proc.
Conference Minutes of human factors in ACM computing system for 32 years, Page2259-2268.
ACM, April 2014. Article (CrossRefLink)Kim, B. J.
, \"Develop an intelligent medicine box to improve the treatment compliance of patients who need long-term treatment
Journal of Korea Institute of Electronic Communications Science, Volume 18, no. 4, 611-617, 2013. Article(CrossRefLink)Chang, W. W. , Sung, T. J. , Huang, H. W. , Hsu, W. C. , Kuo, C. W. , Chang, J. J. ,. . . & Yang, Y. J.
\"Intelligent drug system using Wireless Sensor Network Technology\", sensors and actuators172, no. 1, 315-321, 2011. Article (CrossRefLink)Axisa, F. , Schmitt, P. M. , Gehin, C. , Delhomme, G. , McAdams,E. , & Dittmar, A.
, \"Flexible technology and smart clothing for citizenship medicine, home healthcare and disease prevention\", IEEE biomedical information technology deal, Vol. 9, no. 3, 325-336, 2005. Article (CrossRefLink)Coyle, S. , Lau, K. T. , Moyna, N. , O\'Gorman, D. , Diamond,D.
Francisco, F. ,. . . & Taccini, N. , \"BIOTEX--
Biosensing textiles for personalized medical management, \"IEEE Biomedical Information Technology, Volume 114, no. 2, 364-370, 2010. Article (CrossRefLink)Grillet, A. , Kinet, D. , Witt, J. , Schukar, M. , Krebber, K. ,Pirotte, F. , & Depre, A.
, \"Fiber optic sensors embedded in medical textiles monitored by healthcare\", IEEE Journal of sensors, Vol. 8, no. 7, 1215-1222, 2008. Article (CrossRefLink)Lopez, G. , Custodio, V. , & Moreno, J. I. , \"LOBIN:E-
Textiles and Wirelesssensor-network-
Based on the health care monitoring platform in the future hospital environment, \"IEEE Transaction of biomedical information technology, Volume 114, no. 6, 1446-1458, 2010. Article (CrossRefLink)Paradiso, R. , Loriga, G. , Taccini, N. , Gemignani, A.
And Ghelarducci, B. , \"WEALTHY-
Wearable medical system: a new frontier of electronic medical care
Textile, Journal of telecom and information technology, 105-113, 2005. Article (CrossRefLink)Pacelli, M. , Loriga, G. , Taccini, N. And heaven, R.
, \"Sensing fabric for monitoring physiological and biochemical variables: E-
\"Textile solutions\" in Textile \".
Medical devices and sensors, 2006.
Summer class at rdIEEE/EMBSInternational, pp. 1-4.
IEEE, No. Article (CrossRefLink)George, B. , Zangl, H.
, Bretterklieber, T. And brasur, G.
\"Induction of a combination
Capacitive proximity sensors for seat occupancy detection, \"IEEE Transactions on instruments and measurements, Volume 159, no. 5, 1463-1470, 2010. Article (CrossRefLink)Meyer, J. , Arnrich, B. , Schumm, J. , & Troster, G.
, \"Design and modeling of textile pressure sensors for sitting position Classification\", IEEE Journal of sensors, Vol. 10, no. 8,1391-1398, 2010. Article (CrossRefLink)Xu, W. , Huang, M. C. , Amini, N. , He, L.
, And sarrafzadeeh, M. (2013).
\"Ecushion: textile pressure sensor array design and calibration for sitting posture analysis,\" IEEE Journal of sensors, Vol. 13, no. 10, 3926-3934. Article (CrossRefLink)Menguc, Y. , Park, Y. L. , Pei, H. , Vogt, D. , Aubin, P. M. ,Winchell, E. ,. . . & Walsh, C. J.
, \"Wearable soft measurement for gait measurement of human body\", International Journal of Robotics Research, Volume 133, no. 14, 1748-1764, 2014. Article (CrossRefLink)Tirosh, O. , Begg, R. , Passmore, E. , & Knopp-Steinberg, N.
\"Wearable Textile sensor socks for gait analysis\" in Proc \".
Sensing Technology (ICST)
2013 session of the Seventh International Conference618-622.
IEEE, 12 months. Article (CrossRefLink)Mattmann, C. , Amft, O. , Harms, H. , Troster, G. , & Clemens,F.
In Proc, \"Identification of upper body posture using textile filter sensors \".
2007 ieee International Seminar, page. 29-36.
IEEE, 10 months. Article(CrossRefLink)Enokibori, Y.
, & Mase, K, \"with e-
\"Textile sensor\" in Textile \".
Minutes of 2014 ACM International Seminar on Wearable Computers129-
September. ACM. Article (CrossRefLink)Meyer, J. , Lukowicz, P.
\"Textilepressure sensor for muscle activity and motion detection\" in Proc \".
2006 IEEE International Seminar, page. 69-
October 2006. IEEE. Article (CrossRefLink)Holleczek, T. , Ruegg, A. , Harms, H.
\"Textile pressure sensor for motion application\" in Proc \".
Sensors, 2010 IEEE, pp. 732-
In November, 737 per cent, 2010 per cent. IEEE. Article(CrossRefLink)Kim, S. , Leonhardt, S. Zimmerman, N. , Kranen, P. , Kensche,D. , Muller, E.
\"Impact of contact pressure and humidity on signal quality of newly developed textile ECG sensor\" in Qui \".
Medical devices and sensors, 2008. ISSS-MDBS2008.
5 th International Summer School and seminar256-
259, IEEE, June 2008. Article (CrossRefLink)Catrysse, M. , Puers, R. , Hertleer, C.
Van langhenhof, Los Angeles, VanEgmond, H.
, & Matthys, \"integrating texture sensors in a wireless monitoring set\", Sensor and Actuator a: Physics, Volume 1114, no. 2, 302-311, 2004. Article (CrossRefLink)Huang, C. T. , Tang, C. F. , & Shen, C. L.
, \"A wearable textile used to monitor breathing, using yarn-
Based on the sensor, \"inProc.
2006 IEEE International Symposium, 10 months. IEEE. Article (CrossRefLink)Eren, H. , & Sandor,L. D. , \"Fringe-
In Proc, the effect capacitance for tamper-proof housing approaches the sensor.
Industrial Conference sensors, pp. 22-
February 2005. IEEE. Article (CrossRefLink)Ho, J. G. , Wang, C. , & Min, S.
D, \"study on drug compliance using textile proximity sensors\", trade of Korea Association of Electrical Engineers, Volume 165,no. 7, 1257-1262, 2016. Article (CrossRefLink)Lawrence, I.
, & Lin, K, \"consistency correlation coefficient for evaluation of repeatability\", Biometrics, 255-268,1989. Article (CrossRefLink)Jong Gab Ho (1)
And Se Dongmin (1*)(1)
Department of Medical Information Technology, Chunxian University, Ashan, South Korea [e-
Postage: hogabi1988 @ gmail.
Com, sedongmin @ schac. kr](*)
Correspondent: Se Dongmin was received on September 30, 2017;
Accepted on December 28, 2017;
Publisher February 28, 2018 shows the preliminary version of this article on APIC
IST 2017 and was selected as an excellent paper.
The study was supported by the Soonchunhyang University Research Fund and the National Research Foundation\'s biological and medical technology development project (NRF)
Funded by the Department of Science, Information and Communications Technology and Future Planning (NRF-
Zhong Jia Ho received B. S.
2015, degree in medical information technology Engineering, University of Soonchunhyang.
He is now pursuing a master\'s degree in medical information technology at Chunguang University.
His current research interests include biomedical signal processing, medical sensor applications, and pattern recognition.
Se Dongmin received M. S. and Ph. D.
In 2004, a degree in Electrical and Electronic Engineering, Department of Electrical and Electronic Engineering, Yonsei University, Seoul, was 2010.
He is currently an assistant professor in the Department of Medical it engineering at the University of Ashan Soonchunhyang, South Korea.
His research areas include biomedical signal processing, medical sensor applications, and mobile medical technology.