Ijret - Design and Development of Fall Detector Using Fall Acceleration

Published on February 2017 | Categories: Documents | Downloads: 20 | Comments: 0 | Views: 84
of 5
Download PDF   Embed   Report

Comments

Content

 

IJRET: International Journal of Research in Engineering and Technology 

eISSN: 2319-1163 | pISSN: 2321-7308 

DESIGN AND DEVELOPMENT OF FALL DETECTOR USING FALL ACCELERATION 1 1

2

3

4

Sudarshan3B G , Raveendra Hegde , Prasanna Kumar S C , Satyanarayana B S   4

 Assistant Professor, Professor, Professor and Head, Principal, Dept. of Instrumentati I nstrumentation on Technology, RVCE, Bangalore, Karnataka, India,  [email protected]  sudharshanbg@gmai l.com , prasannakumar  [email protected] @rvce.edu.in , principal@rvce  [email protected] .edu.in

Abstract Fall of patients and aged people may become fatal if unnoticed in time. The concept is to have a fall detection system which sends alarm to the concerned people or to the doctor, at the time of eventuality. To minimize fall and its related injuries continuous surveillance of subjects who are diseased and prone to fall is necessary. The article discusses the design and development of a  prototype of an el electronic ectronic gadget which is used to detect fa fall ll among el elderly derly and the patients who a are re prone to it. In this article, the body posture is derived from change of acceleration in three axes, which is measured using triaxial accelerometer (adxl335). The sensor is placed on the lumbar region to study the tilt angle. The acceleration values in each axis are compared twice with threshold and also a delay of 20 secs between two comparisons, to reduce the false alarms. Values of the threshold voltage are selected by experimental methods. The algorithm is executed by microcontroller (PIC16F877A). The location of fall is determined by GPS receiver, which is programmed to track the subject continuously. On detection of fall, the device sends a text message through GSM modem, and communicates it to computer through ZigBee transceivers. The device can also be switched to only alarm if text message is not required. The prototype developed is tested on many subjects and also on volunteers who simulated fall. Out of 50 trials 96% of accuracy is achieved with zero false alarms for daily activities like jogging, skipping, walking on stairs, and picking up objects.

 Index Terms: Fall Detector, Medical Alarming System, Personal Emergency Response System, triaxial accelerometer, microcontroller -----------------------------------------------------------------------***-----------------------------------------------------------------------

1. INTRODUCTION Fall is a clinical feature of many diseases such as Parkinson’s disease, ectopic of heart, vestibulocochlear defects etc. Falls occur even at home and also in hospitals. Increase in number of patients results in difficulties of manual monitoring by the hospital staff, which leave the patients vulnerable to fall. With the advent of modernization and western culture, the nuclear families are on rise. This has led to single aged people living alone with geriatric problems. Falls are usually fatal due to head injuries and also because of not being treated in time. Falls not only cause physical, but also psychological trauma. People with history of fall are more prone to such attacks. The falls result in sustained psychological effects such as fear, increasing dependence [1]. Importance of preventing complications of fall lies in early detection and prevention of fall. According to survey done by Centres for Disease Control and Prevention (CDC), 33% of aged people p eople fall every year [2]. The old-age dependency ratio (the number of people 65 and over relative to those between 15 and 64) is projected to increase from value of 22% in 2010 to 37% by 2050 [3]. Fall related injuries are not only social burden but also economic aspect. Based on data from a survey done in US in the year 2000, total annual estimated costs were between $16 billion and $19 billion for nonfatal fall-related injuries and

approximately $170 million dollars [4]. Fall-related death rates in the United States increased between 1999 and 2004, from 29 to 41 per 100,000 population [4]. In Indian population such a statistics is not available. A low cost personal emergency response system can also be used by fireman and mountaineers who are at the risk of fall and its related injuries. With the above mentioned factors such as medical and economic issues, lies the importance in design and development of a fall detector. Efforts to detect fall among elderly is being done over several years. Detailed literature review gives us four types of approaches to detect detect fall, na namely: mely: wearable sensor, amb ambient ient sensor, combination of wearable and ambient sensor and image processing. The body wearable method is found economical and suitable for both indoor and outdoor scenarios. S.Y.Sim et al. [1] tried to place accelerometer in shoe and experiments shown that the algorithm is sensitive (81.5%) when the sensor is placed in tongue of shoe. WenChang Cheng et al. [5] used chest or waist worn triaxial accelerometer to derive body posture. Cascade Ada-Boost support vector machine is employed to classify fall from other activities with high accuracy (more than 98%). A new approach by placing the sensors on garment is done by Khalil

__________________________________________________________________________________________ Volume: 02 Issue: 09 | Sep-2013, Available @ http://www.i http://www.ijret.org jret.org 57 

 

IJR nter ern nati tio onal Jo Journ rna al of of Res Resee rch in Engineering and Technology  IJRET: Int

eISSN: 2319-1163 | pISSN: 2321-7308 

Niazma Nia zmand nd et aal. l. [6]; [6]; it it can b bee worn worn wit witho ho t any discomfort. The algorithm achieved sensitivity of 97.5% and specificity of 96.92%. The combination of accelerometer and gyroscope is placed on thi thigh and chest region and and the al orithm employed succee suc ceeded ded to give give 92% accura accuracy cy [7]. A n w algorithm was proposed by Ravi Narasimhan [8]. D Daata is a quired by triaxial computed by accel accelerome erometer tervalue ands. The threshold threshold window win sensitivity experi exp erimen mental tal values. spe specif cific icity ity dow of 10 is% and of 99% was achieved by placing the accel rometer on torso an and d by app pply lyin ing g alg algor oriith thm m. Ma Many ny rese reseaa chers worked to develop fall detecting system using built-in tri-axial accelerometer of mobile phone. Frank posaro et al.[9] developed an application for android phone. It asks the user to comm commun unic icat atee wh when en a fa fall ll is dete detect cted ed,, mes messages are sent to sociall contacts socia contacts if not repl replied. ied. The emergenc emergenc alarm is raised if both fail. Yi He et al. [10] considered smar phone as a waist worn devi device ce and d deve evelope loped d an algori algorith th to send MMS (multimed (mult imedia ia message message service) to pre-selec pre-selecte te contacts and the location locat ion is de determ termined ined by GP GPS S coordinate coordinate and Google map. det ector Fig -1: Block Diagram of fall detector The proposed method in this article is to etect falls which inc includ ludee only only cchang hangee of pl plane ane of human human bo y due to various reasons such as geriatric problems and its associated diseases like Parkinson’s disease, ectopics of heart, vestibulocochlear defects etc.Derived human body posture is compared to predefined threshold values to separate fall from daily acti activi viti ties es.. R Rep epea eate ted d rrea eadi ding ngss of of sen senso sorr res resp ponse for different til iltt aang ngle less giv givee tthr hreesh shol old d val value ues. s. Si Sinc ncee, c ange of plane of body and sudden sudden changes changes in accelerat acceleration ion are involved in fall; triaxial accelerometer is a suitable sensor. The changes of acceleration in 3 axes are monitored continuously. False alarmss are reduc alarm reduced ed by deriving deriving posture posture twice separated by 20 second delay. The section 2 of of the paper describes me methodology to detect fall fall.. S Sel elec ecti tion on of thre thresh shold old vo volt ltag agee is is dis discu cussed in section 3. The suita suitable ble ana anatom tomica icall pos positi ition on to pla pla e the device is expl explai aine ned d in in sect sectio ion n 4. 4. Resu Result ltss of expe experi rim ments are given in sectio sec tion n 5. The The pap paper er is is co concl nclude uded d in the the secti section 6.

2. METHODOLOGY TO DETECT F LL The The ttec echn hniq ique ue of dete detect ctin ing g ffal alll reli relies es on der deriving human body po possture ture wit with a su suiita tab ble se sens nsor or plac placeed at at app pprr priate anatomical po possiti tion on and an eff effecti ectiv ve algo algorrith thm m whic which hp prr cisely distinguish daily activities and fall. Since, change of lane of body and sudden sud den chan changes ges in in acc accele elerat ration ion are are invo involv lv d in fall; triaxial accele acc elerom romete eterr is is a suitab suitable le sensor sensor.. The The ch chan an es of acceleration in 3 axes axes are moni monitore tored d contin continuou uously sly.. Wh n the changes in ac acce cele lera rati tion on fa fall ll in the the w win indo dow w of th thre ress old values, it is decided as fall.

The block diagram (Fig -1) possesses four main sections, input, the controller, co munication protocols and output devices. device s. The The iinpu nputt co consi nsissts of acceleration values and GPS data. The transmitter pin f GPS is connected to the receiver (UART) (UA RT) o off mic microc rocont ontrol rolle ler. Location data is read serially byte by byte which gives the in ormation of longitude and latitude. The triaxial accelerometer sensor (adxl335) is used to derive body posture of the subject. Acceleration and angle information of three axes is produced as three analog signals whic which hv var ary y with with bo bod dy po post sture. Acceleration value generated in each axis is read through separate pins, selecting one analog input at a time. Analog signals generated by the sensor are di digi giti tize zed d by analo analog g tto od dig igiital converter of the microcontroller (PIC16F877A). The di itized values are compared to pr pred edef efin ined ed thr thres esho hold ld va valu lu s. Human body posture is derived twic twice; e; betw betwee een n two two po post stu u e readings a delay of 20 seconds is introduced. The delay time helps to reduce false alarms. ZigBee Zig Bee tra transc nsceiv eiver er pair (Tarang F4) is used for comm commun unic icat atio ion nb bet etwe ween en t e microcontroller and the computer and indicators. An alert essage is sent to a hospital phone number when a fall is is de detected. The GSM modem used is SIM 300 V_7.03. ZigBee tr nsceiver and GSM modem are connec con nected ted to R RS23 S232 2 po po t of microcontroller board via a switch. switc h. Th Thee sy system stem can b switched send either SMS alert or  just a ZigBee alert to the indicators (LED and beeper). This facilit y avoids sending unnecessary SMS while patient is at homee w hom with ith fam family ily mem member bers. The mic microc rocont ontrol roller ler boar boar , the ZigBee transceiver and the GSM modem modem aare re power poweree up by a single DC source of 12V, 2A. The The ssens ensor or bo boar ard d an GPS device use 5V generated by builtt in voltag buil voltagee regula regulator tor of microcontroller board.

______________________________ ___________________________________ ________________________ Volu Vo lume me:: 02 02 Iss Issu ue: 09 | Se Sepp-20 2013 13,,  Ava Avail ilable @ http://www.i http://www.ijret.org jret.org 58 

 

IJR nter ern nati tio onal Jo Journ rna al of of Res Resee rch in Engineering and Technology  IJRET: Int

eISSN: 2319-1163 | pISSN: 2321-7308 

2.1 Flow Chart for the Algorithm

2.2 Algorithm

The Fig -2 depicts the detailed flow f program. The microcontr micro controller oller is programm programmed ed to to monit monit r and track the subject subje ct conti continuous nuously. ly. The init initializ ializatio atio part includes configurin confi guring g analog to digital conv converte erte and Universal Asynchronous Sy Synchronous Receiver Tra smitter (UART). To remove false alarms induced by fall ike activities the ac acce cele lera rati tion on valu values es ar aree meas measur ured ed an and d co com mpared two times. Betwee Bet ween n two two m mea easure sureme ments nts and compar compariso isons, a delay of 20 second sec ondss iiss intr introdu oduced ced.. The flo flow w chart chart shows two modes of operation, mode1: send text message to pre-stored phone numbers, mode2: send only alarming signal to the indicators. Mode 2 is used when the patient is alone and the device is switch swi tched ed to mode mode 1 when when famil family y me membe mbers rs a e around.

microcontroller er Step1. Initialize serial communication ports of microcontroll Step2. Ste p2. Conf Configu igure re ADC ADC an analog input channel Step3. Initialize GPS and SM modules. Step Step4. 4. Re Rece ceiv ivee an anal alog og inp inputs from sensor. Step5. Receive location in ormation from GPS. Steep6 St p6.. Com Compa pare re the the d dig igital values of sensor signal with predefined thresholds. If acceleration is greater than the threshold go to step7, else go to step4. Step Step7. 7. Wa Wait it fo forr ttim imee t and again read acceleration values. Comp Compar aree wi with th sa same me thre thresh shold again. Step8. Is fall detected? If es go to step9, go to step4 if not. Step9. Send text message to stored numbers, send alarming signal to indicators if he operating mode1. Send only alarming signals to the indicators if mode2.

3. SELECTION OF T RESHOLD VOLTAGE The thresh threshold old voltag voltagee i selected by readings of sensor resp respon onse sess fo forr diff differ eren entt til tilted positions. Fig -3 shows sensor outputs for different fall scenarios. It is observed that for different types of falls.

f or implemented algorithm Fig -2: Flow chart for

 Read adin ing gs of triaxial accelerometer Fig -3: Re

______________________________ ___________________________________ ________________________ Volu Vo lume me:: 02 02 Iss Issu ue: 09 | Se Sepp-20 2013 13,,  Ava Avail ilable @ http://www.i http://www.ijret.org jret.org 59 

 

IJR nter ern nati tio onal Jo Journ rna al of of Res Resee rch in Engineering and Technology  IJRET: Int

Backward Backwa rd fall fall iiss ind indica icated ted by Z aaxis xis,, forw forward and right side fall is indicated by Y axis values, and Z axis indicates left side fal fall. l. From From the the obser observat vation ion of of se senso nsorr respo respons nses, 1.96V, 1.63V, 1.98V are considered as threshold values for X, Y, Z axes respectively.

4. PLAC PLACEM EMENT ENT OF THE THE DEVIC DEVIC BODY

ON HUMAN

eISSN: 2319-1163 | pISSN: 2321-7308 

5. RESULTS AND DI CUSSIONS The compone onents are encl sed in a plastic box and tied in the waist region as shown in Fig -5. The prototype is tested on fiv ivee vol volu unte nteer erss ((he heaalthy lthy adult males) in the PG research lab and RVCE health center. ach subject is made to fall on bench and bed, and many fall like activities (skipping, jogging, pi pick ckiing ob obje jeccts fr from the the floor and walking on staircase). The following follo wing figur figuree show showss on of the subjects simulating fall.

Anatomical position Anatomical position of the device influence influence the accuracy and specificity to a great extent. If the detection lgorithm depends mainly mai nly on the b body ody posture posture and and tilt, tilt, then then tor tor o is more suitable place [11]. Stefano Abbate et al. [11] listed different possible anatomical anato mical positions positions to de derive rive vario various us po postur stures (table 1). Table -1: D  Diifferent positions of sensor corr sponding derived posture

Sensor position

Identified posture

Chest

(standing or sitting), (b nding or lying)

Waist

(bending or standing or sitting), (lying)

Chest Che st + Thigh Thigh

bend bending ing,, lyi lying, ng, sta standi ndin n , sitting

Since algorithm of the project depends on t resholds of tilt, it is suitable to select waist portion to place the device. If the device dev ice is p plac laced ed just just bel below ow stom stomach ach,, obe obessity of the subject may indu induce ce som some til tilt. t. To avo avoid id th this is,, llu u bar vertebrae is appr approp opri riat atee p pos osit itio ion n to to plac placee tthe he dev devic icee aass shown in the Fig -4.

Fig -5: On  Onee of of th th volunteers simulating fall

Out of fifty fifty tria trials ls (te (ten n tria trialls each subject) only two events are not detected. The trials in luded forward fall, falling sideways and backward fall. The missed detection happened when the subjects subje cts knel kneltt down down sl slowl owl and leaned forward without giving any jerk. Since most of th falls not likely to happen this way, the miss missed ed de detec tectio tions ns d not have significance. The fall simulating activities like jogging, skipping, walking on stairs and picking objects did not create any false alarms. The outcom out comee of the exp experi erimen ments is listed in the table 2. As we see from the table table the accurac accurac is 96%.  An nalysis of prototype testing Table -2: A Subject 1 2 3 4 5

Number of

Number

trials 10 10 10 10 10

detections 9 10 10 9 10

of

Number

of

false alarms 0 0 0 0 0

The snaps snapshot hotss of assembl assembly of components are shown in the Fig -6.

Fig -4:  Placement of the device.

 

______________________________ ___________________________________ ________________________ Volu Vo lume me:: 02 02 Iss Issu ue: 09 | Se Sepp-20 2013 13,,  Ava Avail ilable @ http://www.i http://www.ijret.org jret.org 60 

 

IJR nter ern nati tio onal Jo Journ rna al of of Res Resee rch in Engineering and Technology  IJRET: Int

[6] 

 

eISSN: 2319-1163 | pISSN: 2321-7308 

Khalil Niazmand, Claudius Jehle, Lorenzo T. D Angelo an and d Tim Tim C. Luet Lueth, ‘A New Washable Low-Cost Garment fo for E Ev veryday Fall Detection’, 32nd Annual International C Co onference of the IEEE EMBS Buenos Aires, Argentina, ugust 31 - September 4, 2010. PP 6377-6380.

[7]

Fig-6: Assembly of components

 

CONCLUSIONS In this this re resea search rch pap paper, er, we had ach achiev ieved ed o r primary goal of creating creat ing a wor working king prototype prototype able to rrecogn ecogniize both dangerous posture postu re and and falls falls from from non-fa non-falls, lls, with w wirele ireless communication to the indicators and computer. Looking at the underlying det detect ection ion proc process ess,, our our fall fall det detect ection ion syst system improves on previous systems and designs by giving ero false alarms, bearing low cost, cost, and with new new anatomical position for the sensor. or. W Wee inc incorpor porated hy hybrid fa fall de de ection algorithm derived fr from eex xisting al algorithms, aan nd in in erfaced the GPS rece re ceiv iver er suc succe cess ssfu full lly y tto o loca locate te ffal all. l. The The acc accu u acy is 96%.

REFERENCES [1] 

[2] 

[3] 

[4]  [5] 

S.Y.Sim, H. H.S.Jeon, G.S.Chung, S S.. .Kim, S.J.Kwon, W.K.Lee, K.S.Park, ‘Fall detection algorithm for the elderly using acceleration sensors on the shoes’. 33rd Annual International Conference of the IEEE EMBS Boston, Massachusetts USA, August 30 - September 3, 2011. pp 4935-4938. GohYo GohYong ngli li,, O Ooi oi Sh Shih ih Yin Yin aand nd Pa Pang ng Ying Han, ‘State of the Art: A Study on Fall Detection’, orld Academy of

Qiang Jo John hn Mark Hanson, Adam Bar arth th,, ‘Ac ‘Li, Accu cura rate te,, A F . stStankovic, Fall Detection Using Gyroscopes and Accelerometer-Derived Posture In Inffor orma mattion on’’Wear Wearaa le and Implantable Body Sensor Networks,. BSN 2 09. Sixth International Workshop, 2009 Digita Digitall Object Identifier: 10.1109/BSN. Pp 138143. [8]  Ravi  Narasimhan, ‘Skin-Contact Sensor for Automatic Fa Fall ll Dete Detect ction ion’, ’, 34 h Annual International Conference of the the IIEE EEE E E EMB MBS San Diego, California USA, 28 August - 1 September,2012. pp 4038-4041. [9]  Frank Sposaro and Gary Tyson, ‘i Fall: An android Applic App licat ation ion for Fa l Monitoring and Response’, 31st An Annua nuall Inte Intern rnat atio ion n l Conference of the IEEE EMBS Minneapolis, Minn sota, USA, September 2-6, 2009. Pp 6119-6122. [10]  Yi He, Ye Li, and Shu-OiBao, ‘Fall Detection by BuiltIn Tr Trii-Ac Acce cele lero rome metter of Smartphone’, Proceedings of the IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI 2012) Hong Kong and and Shenzhen Shenzhen, China, 2-7 Jan 2012. Pp 184-187. [11]  Stefano Abbate , rcoAvvenuti , Guglielmo Cola , Paolo Cors “Rec gnition of false alarms in fall detection systems” 30th Annual International IEEE EMBS EMB S Con Confer ferenc enc Vancouver, British Columbia, Canada, August 20-24, 2008. pp 234-239.

Science, Scienc e, Engi Engineer neering ing and Technolo Technolog g 62, pp 294-298, 2012. Kora Koray yOzca Ozcan, n, Anvi Anvith thKa Kattte teMa Maha haba ballagiri, Mauricio Ca Casa sarres es,, Membe emberr, and Sene Senem mVeli Velipa pasalar, ‘Automatic Fall Detection and Activity Classification by a Wearable Embedded Smart Camera’, IEEE Journal on Emer Em ergi ging ng an and d Sele Select cted ed Topi Topiccs in Ci Cirr uits and Systems, VOL. 3, NO. 2, JUNE 2013. Pp 125-1 37. Lean Leanne ne C Cur urrie rie,, ‘Pati ‘Patien entt Sa Safe fety ty and Quality: An Evidence-Based Handbook for Nurse ’, chapter 10. Fall an and d Inju Injury ry Prev Preven enti tion on.. AHRQ AHRQ P Pub ubli lica cation No. 08-0043. Wen-Chang Wen-C hang Cheng and DingDing-Ma Ma Jhan, ‘Triaxial Accelerometer-Based Fall DetectionMethod Using a Self-Constructing Cascade-AdaBoost-SVM Classifier’, IE IEEE EE Jo Jour urna nall of Bi Biom omed edic ical al an and d Health Informatics, VOL. 17, NO. 2, MARCH 2013, pp 410-419.

______________________________ ___________________________________ ________________________ Volu Vo lume me:: 02 02 Iss Issu ue: 09 | Se Sepp-20 2013 13,,  Ava Avail ilable @ http://www.i http://www.ijret.org jret.org 61 

Sponsor Documents

Or use your account on DocShare.tips

Hide

Forgot your password?

Or register your new account on DocShare.tips

Hide

Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in

Close