International Journal of Modeling and Optimization, Vol. 2, No.

1, February 2012

Proposed Design of a Low Cost mHealth System

Rajat Kumar and Arvind Rehalia

Abstractthis paper proposes a model for low cost design of mhealth solutions. The model presented here is based on RF transmitter and receiver, microcontroller and cellular network integration and uses F-BUS protocol for transmitting and receiving data using a NOKIA 3310 handset. The proposed design aims at eliminating the need of smart-phone dependent costly mhealth solutions. Also, the current paper presents the essential comparative data and statistics to support the concerned proposed design. Index TermsBio-sensors, f-bus protocol, mhealth.

discusses the proposed low cost design for mhealth solutions. Section IV gets into the core of the F-BUS protocol for NOKIA handsets. Section V discusses the merits of the proposed design and finally Section VI concludes the paper along with a analysis presenting future scope and the challenges for mhealth system.

II. RELATED WORK Different universities and institutions are undertaking mhealth research projects at a very fast pace and the developments across this field are at full swing. A large number of research papers have been published by researchers all across the globe concerning mhealth solutions. The fact that , a majority of population today uses a mobile phone strengthens the belief in implementation of such a technology as shown in Fig. 1.

I. INTRODUCTION Mobile phones have made life simpler by bridging the communication gap among different sections of developing economies. Today, low cost mobile phones have become omnipresent and continuous developments in this sector have enabled various other ideas and technologies to fuse together. Mobiles no longer are used for calling purposes only and have been successful in establishing their market in areas like mobile gaming,navigation,multimedia applications etc. The increased potential of mobile phones have thus enabled the bio-instrumentation community to develop and implement mobile technology for health care service delivery. The primary goal of such a developmental project aims at improving safety and reducing overall costs. mhealth refers to the delivery of health care facilities supported by mobile devices using cellular,blue-tooth or wireless networks. mhealth system uses the portable biomedical equipments and has potential for providing low cost medical services for everyone from anywhere at anytime. The current paper proposes a low cost design of a mhealth system for continuous patient monitoring and data logging. According to the author in [1] , rising costs of health care service delivery is a menace and needs to be dealt with proper measures that reduce the cost of medical services and at the same time preserve the quality of treatment. Therefore, special care has been taken while designing the model of the proposed design in preserving simplicity,cost effectiveness and security. This paper presents different sections of the proposed design independently. Section II presents the related work done earlier and a case study. Section III
Manuscript received November 8, 2011; revised January 14, 2012. Rajat Kumar is with the Bharati Vidyapeeth's College of Engineering, Guru Gobind Singh Indraprastha University , New Delhi , India (e-mail: rajatk489@gmail.com). Arvind Rehalia is with the Bharati Vidyapeeth's College of Engineering, Guru Gobind Singh Indraprastha University, New Delhi, India (e-mail: rehaliaarvind@gmail.com).

Fig. 1. Infrastructural statistics (in millions) [2].

CASE STUDY: Child-Count [3] : This program was started in July 2009 in Sauri, Kenya for monitoring more than 9500 children, under five years of age, by using mobile application based on Rapidsms ( a free and open-source platform for data collection and communication using short messaging service of mobile phones) by 100 community health workers. The registration statistics on daily basis has been shown in fig.2. The major goals of this project were : 1. Identification of risk of acute malnutrition by measuring a child's mid upper arm circumference. 2. Home based testing for malaria. 3. Home based treatment of children with diarrheal illness using ORS and Zinc supplements. In this project, community health workers used SMS messages to register a patient and sent in their data. The major challenges that came up were as under : 1. Community health workers required additional training for using the system. 2. Several phones went missing and few needed
15

International Journal of Modeling and Optimization, Vol. 2, No. 1, February 2012

replacement. This problem can be overcome by using cheap, entry-level phones. 3. About 10% of the messages were rejected due to improper formatting by community health workers.

Fig. 2. New child registrations [3].

Apart from this project, there are many other related works going on. Table 1 shows few such mhealth initiatives taken by different institutions all across the globe.
TABLE 1: EXAMPLES OF MHEALTH INITIATIVES [4]

3. Micro-controller 4. Nokia handset 3310 5. EEPROM 6. Signal Conditioning Unit A bio-sensor is an analytic device incorporating a deliberate and intimate combination of a specific biological element (that creates a recognition event) and a physical element (that transduces the recognition event)[5]. Bio-sensors are used for blood pressure monitoring, pulse measurement, ECG sensing etc. A specific bio-element such as an enzyme, nucleic acid etc. recognizes a specific analyte under study and then the transducer element transforms the signal resulting from the interaction of bio-element and the analyte to useful electrical signals. In the current design, bio-sensors are placed on to the body and transducers are interfaced along with them. This system is further attached to a small RF transmitter. Therefore, the system thus constitutes a Body-Area Network (BAN) that allows the patient to be free from any wired monitoring connections and the BAN becomes a roaming Unit. Within a range of around 175m, a central unit is placed which has a RF receiver that receives the data and sends it to the signal conditioning unit. After signal conditioning, the analog to digital conversion is done and the digital input is given to the microcontroller. The microcontroller is attached to a GSM phone with GPRS service. In this case, NOKIA 3310 has been chosen for the required study. Most NOKIA phones have F-BUS connection that can be used to connect a phone to a microcontroller. These phones are cheap and entry level in nature. The F-BUS protocol will allow us to send and receive SMS messages [6]. Apart from this, the micro-controller is also attached to a EEPROM that can be used as a local storage for maintaining a data base. Timer programming can be done in the microcontroller to continuously store data in EEPROM after fixed intervals using serial communication. The whole process or the proposed design can be easily understood with the help of the fig. 3.

It was analyzed that most of the previous work done on mhealth using cellular phones had the following similarities : 1. Use of smart-phones based on Operating Systems such as iOS, Android, Windows or Symbian^3. 2. Use of Java or C# based application development platform. 3. Separate application design for monitoring and data collection. 4. Use of Blue-tooth technology for constructing a body area network or a roaming unit for the mhealth system.

Fig. 3.Figure explaining the design of central unit.

III. PROPOSED LOW COST DESIGN FOR MHEALTH The proposed model uses the following important components: 1. Bio-sensors 2. RF transmitter and receiver
16

After the data has been transmitted from the central unit i.e. patient's mobile, it reaches the secondary unit. The secondary unit has installed monitoring and recording devices and the patient's continuous monitoring can be done through these devices. The patient's mobile sends the SMS only after a particular time (timed using microcontroller) or when the

International Journal of Modeling and Optimization, Vol. 2, No. 1, February 2012

secondary unit asks for data. The figure (fig. 4) describes the whole process. Also, the coding can be done such that when the secondary unit requires previously stored data, the central unit responds by sending the data stored in the EEPROM.

Fig. 4. Figure explaining the design of secondary unit.

IV. NOKIA F-BUS FRAME FORMAT Most NOKIA phones provide the facility of F-BUS for connecting to a PC or microcontroller using serial port. The phones under this category are cheap and entry-level and don't include feature of USB port for communication. The pin configuration of NOIKA 3310 is as in fig. 5. Apart from F-BUS protocol Nokia has MBUS protocol also. The MBUS protocol is a one pin, bi-directional bus for data transfer at a baud rate of 9600 bps and is half duplex in nature. It has 8 data bits, odd parity and one stop bit. The F-BUS protocol, however, is more advanced and has one pin for transmitting and one pin for receiving the data. It has a baud rate of 115,200 bps and is full duplex in nature. It has 8 data bits, no parity and one stop bit.

SMS HEADER Byte 6 to 8: Start of SMS Frame Header Byte 9 to 11: Send SMS Message (0x01,0x02,0x00) Byte 12: SMS Centre number length , 0x0a Byte 13: SMSC number type , 0x81: unknown, 0x91: national Byte 14 to 23: SMS centre phone number TRANSFER PROTOCOL DATA UNIT Byte 24: Message Type, 1: sms submit, 0: sms deliver Byte 25: Message Reference if SMS deliver and validity indicator used Byte 26: Protocol ID, 0x00 Byte 27: Data Coding Scheme Byte 28: Message Size (34 bytes long for unpacked message) DESTINATION'S PHONE NUMBER Byte 29: Destination's number length Byte 30: Number type 0x91-international, 0xa1-national Byte 31 to 40: Destination's Phone Number VALIDITY PERIOD Byte 41: Validity-Period Code, 0xFF Byte 42 to 47: Service Centre Time Stamp SMS SUBMIT Byte 48 to 92: SMS message packed into 7 bit characters Byte 93: Always 0x00 F-BUS FRAME ENDING Byte 94: Packet Sequence Number Byte 95: Padding Byte Byte 96 and 97: Odd and even checksum bytes.

V. MERITS OF THE PROPOSED DESIGN There are several factors that make the proposed design well suited for adoption as a mhealth system. This section discusses about some such merits and presents facts and figures supporting the same. They are as under : Cost: It is important to consider the economic input required for the design of a mhealth system that can provide the required services to the people in need. It has been observed that most of the mhealth systems include the use of smart-phones having much greater price range as compared to normal data entry-level mobile phones like Nokia 3310. Also a low cost body area network based on RF technology has been used instead of blue-tooth technology. Considering the earning of people in developing nations, it is necessary to reduce this cost as much as possible. Table 2 reveals the whole truth and expresses the need of low cost mhealth solutions.
TABLE 2: REGIONAL AND GLOBAL POVERTY [7]

Fig. 5. Pin configuration of NOKIA 3310 for serial communication

For the current paper, F-BUS protocol has been analyzed and studied. For synchronizing UART in the phone with microcontroller, a string of 0x55 or 'U' is sent 128 times. The Nokia protocol uses a set of commands that allow the user to make calls and send and receive SMS messages. It is important to note that Nokia uses 7-bit character packing. The full SMS message frame contain the packed SMS and all other details like source address, destination address, message reference etc. The complete FRAME FORMAT[6] is as under : Total Frame : 98 byte F-BUS FRAME HEADER Byte 0: F-BUS Frame ID , 0x1E Byte 1: Destination address , 0x00 Byte 2: Source address , 0x0C Byte 3: Message Type , 0x02 Byte 4: Message Length. Byte 5: message Length
17

International Journal of Modeling and Optimization, Vol. 2, No. 1, February 2012

Data Base Management System: The smart-phones require coding of different applications under JAVA/C# or some other platform for continuously storing the acquired data and also the same application is designed differently for different phones. This problem is solved by using a EEPROM that continuously stores data and releases it as per the need. This requires only one time programming to be done for activating the serial communication between the microcontroller and the mobile phone. Security: Most of the medical institutions use bar code or the biometric technologies for patient identification. Patient's medical data has to be secure and private and independent from any mischievous alternations. For achieving the same with the proposed design, it is possible by using a patient ID which when sent from the secondary unit to the primary unit will be tested and accordingly data will be released through SMS from the EEPROM. This is quite simple to achieve as the microcontroller used in the primary unit can be programmed using conditional statements for processing the request only after receiving the correct patient ID. Apart from all these, the problems of incorrect data entry are also avoided as in this case, the manual entry has been replaced by automated microcontroller based data entry.

mhealth technologies with integration of 4G, multimedia interface, high speed connectivity and response system, lower costs, faster, secure and accurate mhealth designs. Finally the paper concludes that the proposed design of the mhealth system presents easy solution to many of the problems that are faced during implementation of mhealth technology. If the current design fits in with both the medical model and the economic model, then there can be a great help to the people of developing economies and THE THIRD WORLD. REFERENCES
[1] M. E. Porter and E. O. Teisberg, Redefining Health Care: Creating Value Based Competition on Results, 1st ed., Harvard Business School Press, ch. 1 , pp. 17-32. mHealth for Development: The Opportunity of Mobile Technology for Health Care in the Developing World, UN Foundation and Vodafone Foundation, UK , pp. 7. 2009. M. Berg, J. Wariero, and V. Modi, Every Child Counts The use of SMS in Kenya to support the community based management of acute malnutrition and malaria in children under five, Initial Report, ChildCount+, Millenium Villages Project, 2009. J. Black, F. Koch, L. Sonenberg, R. Scheepers, A. Khandoker, and E. Charry et al. Mobile Solutions for Front Line Health Workers in Developing Countries, ehealth Networking, Applications and Services, HealthCom, pp: 88-93 , December 2009. R. Shahriyar, Md. F. Bari, G. Kundu, S. I. Ahamed, and Md. Mostofa Akbar, Intelligent mobile health monitoring systems, International Journal of Control and Automation, vol. 2. no.3, September 2009. Wayne Peacock (July 2010), A Introduction to Nokia F-Bus [online]. Available : www.embedtronics.com/nokia/fbus.html Poverty in Numbers: The changing State from 2005 to 2015, The Brookings Institution, Washington, D.C., USA, January 2011, pp. 4. Barriers and Gaps Affecting mHealth in Low and Middle Income Countries, Earth Institute, Washington, D.C., USA, 2010. L. Bos, D. Carroll, L. Kum, A. Marsh, and L. M. Rao, Future Visions on Biomedicine and Bioinformatics 2, Springer 2011, pp: 11-16.

[2]

[3]

[4]

[5]

VI. FUTURE SCOPE, CHALLENGES AND CONCLUSION This final section of the current paper presents a detailed analysis of the future scope and the challenges for implementation of mhealth system. No-doubt, mhealth is a technology that can bring revolution into the field of medical service delivery but the problems associated with the current implementation of mhealth are important to be considered for improvement and better development design for future. Authors in [8] have undergone a lot of literature review and have identified framework for mhealth under following categories: Treatment Compliance Data collection and disease surveillance Health information system and point-of-care support Health promotion and disease prevention Emergency medical response A close study of the above factors defines the need of analysis of the current mhealth system. Failure of treatment compliance in the areas of poor drug supply, high cost of communication links, limited availability of mobile internet connectivity, poor information access due to bandwidth limitations and limited data transfer rate of the current mobile wireless systems [9] are the factors affecting mhealth model. For mhealth to be successful nationwide or at any other larger level, data collection, storage and exchange should be cost efficient and safe. Response system for emergency needs should be fast and accurate. Drug supply should be regular and efficient at areas of mhealth implementation so that treatment compliance framework is kept in place. The upcoming years ahead will definitely see a rapid growth in

[6] [7] [8] [9]

Rajat Kumar was born on 28th September 1992, is pursuing his B.Tech in Instrumentation and Control Engineering from Bharati Vidyapeeth's College Of Engineering, New Delhi. He is involved in various projects concerning Data Acquisition. His areas of interest are data acquisition, embedded systems, control systems, process control and virtual instrumentation.

Arvind Rehalia received his M.Tech in Process Control And Instrumentation from Kurukshetra University and B.Tech in Electronics And Communication Engineering from Himachal Pradesh University, Shimla. He is presently pursuing his PhD. and is working as an Asst. Professor at Bharati Vidyapeeth's College Of Engineering, New Delhi. His areas of interest are process control , bio-instrumentation , digital image processing and bio-electronics.

18