PERSONAL TELEMETRIC SYSTEM – GUARDIAN

Dalibor Janckulík, Ondřej Krejcar and Jan Martinovič

VŠB – Technical University of Ostrava, FEI, 17. listopadu 15, 708 33 Ostrava, Czech Repulic

Keywords: PDA, embedded device, biotelemetry, wireless, ECG.

Abstract: This project deals with the problems of utilization of mobile equipment working in the biomedicine field,

particularly telemedicine. This field is relatively new; it focuses on the observation of life functions from a

distance. Practically developing system works with an ECG sensor connected to mobile equipment, such as

PDA/Embedded, based on Microsoft Windows operating system. The whole system is based on the

architecture of .NET Compact Framework, and other products, such as SQL Server by Microsoft too. This

work also deals with the communication of mobile equipment with sensors and with the server via

Bluetooth, WiFi, and GPRS/EDGE. The mobile equipment used serves primarily for measuring and

processing of data from the sensors and their visualization as a graph. The data is also given to the server for

further processing and the analysis of current health of the patients, due to small efficiency of the mobile

equipment (Janckulík, 2007). The main task we deal with in the server part of application is receiving of the

data via web services and further processing, management and analysis of this data. For the analysis of

received data and further evaluation of the electrocardiogram, there is a self-organizing neural network

(Vašíček, 2007).

1 MOTIVATION

Many middle-aged people like businessmen, CEOs,

managers and other have very hectic lives with

much stress and without good ways of living.

Sometimes these people have a collapse, breakdown

or heart attack and must be in hospitals or health

resorts for a long time to regenerate their bodies.

The time that they spend in these institutions, is

nonutilisable and very long for them. Possibilities of

today physics are restricted by many of prescripts so

patients cannot use some of the newest techniques

(like hyperbaric or arctic chambers), which make it

possible to reduce the regeneration time by weeks or

months. For example, these chambers are restricted

to patients in the first six months after heart attack

due to no information about patients’ conditions

during the procedure.

Here is the main area of utilization of our

telemetric system. Of course the use of our system is

not limited only to businessmen, but it is targeted to

middle-aged people with some knowledge about

new technology like mobile phones. The price of

client devices of our system is not low, so we

suppose people who can invest to these sorts of

assistants.

The basic idea is to create a system that controls

important information about the state of a

wheelchair-bound person (monitoring of ECG and

pulse in early phases, then other optional values like

temperature or oxidation of blood ...), his situation

in time and place (GPS) and an axis tilt of his body

or wheelchair (2axis accelerometer).

Values are measured with the existing

equipment, which communicates with the module

for processing via Bluetooth wireless

communication technology. Most of the data

(according to heftiness) is processed directly in

PDA or Embedded equipment to a form that is

acceptable for simple visualization. Two variants

are possible in case of embedded equipment – with

visualization and without visualization (entity

with/without LCD display). Data is continually sent

by means of GPRS or WiFi to a server, where it is

being processed and evaluated in detail. Processing

and evaluating on the server consists of - receiving

data, saving data to data storage, visualization in an

advanced form (possibility to recur to the older

graph, zoom on a histogram (graph with historical

trend), copying from the graphs, printing graphs),

automatic evaluation of the critical states with the

help of advanced technologies (algorithms) that use

170

Janckulík D., Krejcar O. and Martinovi

c J. (2008).

PERSONAL TELEMETRIC SYSTEM – GUARDIAN.

In Proceedings of the First International Conference on Biomedical Electronics and Devices, pages 170-173

DOI: 10.5220/0001051901700173

 SciTePress

Artificial intelligence to notify the operator about

the critical state and its archiving.

Application in PDA, Embedded equipment is

comfortable, with minimum time - the first

configuration, but also configuration after downfall

of application. The level of visualization will be

lower. The described system can be used with small

modifications for monitoring of patients in hospitals

or people working in extremely hard conditions. The

biggest limitation is the availability of measuring

devices in acceptable and adaptable sizes or

comfortable enough to have one around.

The measuring device (ECG, plethysmograph)

was tested in extreme conditions in a cryogen room

in Teplice (-136°C), where the final system will be

installed. Implementation of the data transmission

security was not solved. The whole system is

classified as „work in progress“ system and it is in a

testing phase where we found mistakes and repaired

them.

2 SYSTEM ARCHITECTURE

This system consists of several interconnected parts

that can communicate among themselves, so they

can approach their function. These parts are:

measuring sensor, PDA or embedded device, and

server for data processing and evaluation. Beside

these, the most important parts that are worth

mentioning are various kinds of accessories, such as

GPRS, WiFi, GSM, Bluetooth or GPS modules. By

means of them we can communicate. We mostly use

the fastest technology in signal coverage.

Data acquisition and data transmission are the

most important parts of this system. The

Responsible for correct working are: correct sensor

configuration, sensor calibration, data transfer

synchronization, and mutual communication

between sensors and data receiver.

Figure 2: Bluetooth communication.

The system can display saved data from a database

file. The doctor can configure or set a neurone

network. A change of that is shown in the XML file

enshrining, where the neurone network setting of the

patient is kept.

The doctor receives information about

worsening of patient’s status. In case of doctor’s

reaction, he sends for an expert assistance, such as a

helicopter or an ambulance. In case of false alarm,

he can configure a neurone network or leave it

unchanged if that was a sporadic incorrect

interpretation.

The patient can browse data concerning his

health status. Measured data is sent to server by

WEB service.

Client’s data are not only received but also

preprocessed (data checking, risk elimination etc.).

Measured data is saved. And now, it is possible to

analyze the data using a neurone network. If the

analysis shows that the measured data from ECG is

critical, a warning is sent. That notifies a doctor of

incoming data.

2.1 Mobile Part

The main part of the whole system is an Embedded

or PDA device. The difference in applications for

measurement units is the possibility to visualize the

measured data in both Real-time Graph and

Historical Trend Graph, which can be omitted on an

embedded device.

PDA is a much better choice for Personal

Healthcare, where the patient is already healthy and

needs to review his condition, or for multiple person

usage. Embedded devices can be designed for one

user, with the option to use an external display used

for settings or with the possibility of usage in

extreme conditions.

The application is communicating with an ECG

Measurement Unit (Corbelt or Blue Keg) through a

virtual serial port using wireless Bluetooth

technology. Then, after pushing a button, all

necessary parameters are set and the communication

may begin. Measured data is stored on a SD

Figure 1: Measurement schema.

PERSONAL TELEMETRIC SYSTEM – GUARDIAN

171

Memory Card in a database in MS SQL Server 2005

Mobile Edition.

Figure 3: PDA/MDA/SmartPhone visualization.

The performance of available devices seems

insufficient for sequential access; parsing of

incoming packets is heavily time-consuming.

Pseudo paralleling is required. If Windows Mobile

OS versions 2003 to 5.0 are used, the processing of

data from a professional EKG is not realizable due

to thread count limitations. A newer operating

system (Windows Mobile 6) can be used to solve

this.

Figure 4: PDA/MDA/SmartPhone application.

Current application is highly specialized and written

to accommodate specific hardware. Usage of any

other hardware is not possible. This is due to

different methods of packet folding, which are

unique on each device (

Corbelt, BlueEKG datasheet,

2007). This is partly caused by the length of the

Telemedicine branch. Operating of the device is

simplified as much as possible with the least

possible number of steps regarding user registration,

measurement device connection and the

measurement itself. The informations about user, as

ID, name, surname, address and application

properties are stored in the system registry

(HKEY_CURRENT_USER / Software / Guardian).

Working (saving, reading, finding) with registry is

easier and faster as saving this informations in file.

User registry values are crypted with simple

algorithm (shifting char ASCII value).

2.2 Server Part

In order to run a server, an operating system

supporting IIS is needed. IIS is an Internet

Information Server application allowing users to

connect to the web server by the well-known HTTP

protocol. The web service transfers data between the

server and PDA/Embedded devices. It reads the

data, sends acknowledgments, stores the data in the

database and reads it from there. The service is built

upon ASP.NET 2.0 technology. The SOAP protocol

is used for the transport of data, which is in XML

format. That is an advantage since it allows

communication of multiple different technologies

and platforms.

The Wireless ECG approaches a real

professional ECG with data rate as high as 800

records per second (

Corbelt, BlueEKG datasheet,

2007

). That makes 48,000 records per minute and

2,880,000 per hour. Considering 100 patients, the

value gets to 288,000,000 records per hour. Even if

the server accepted only 50 records per second, the

sum of records for 100 patients per hour would be

18 million. That is an extreme load for both the

server and the database system; hence a better way

of storing data is needed.

Methods that devices communicating with the

web service can use include:

 receiving measured data.

 receiving patient data.

 deleting a patient.

 patient data sending.

To observe measured data effectively, visualization

is needed. A type of graph as used in professional

solutions is an ideal solution. To achieve this in a

server application, a freeware Zed Graph library can

be used. For data analysis, neural nets are a

convenient solution. However, there are problems in

the automatic detection of critical states. Every

person has a specific ECG pattern. What is

completely normal for one person can indicate crisis

for another. The Neural net has to learn to

distinguish critical states of each patient separately.

BIODEVICES 2008 - International Conference on Biomedical Electronics and Devices

172

Figure 5: Server visualization.

Basic characteristics of a neuron network:

 10 x 10 neurons.

 learning is based on 3-4 minutes of

recording, which is approximately 36,000-

45,000 recorded values.

 incorrect values are filtered out.

 filtering decreases the amount of values to

about 10%, which is still good enough for

learning.

 the learning cycle with 4,200 values takes

approximately 30 seconds (CPU PIII 1.2

GHz) with C# implementation.

To make the specialist’s or operator’s intervention

possible, the system must be provided with a user-

friendly interface, possibly imitating those on

medical appliances. This area is still in an early

phase of development.

3 DATA ANALYSIS

Data acquired from the measuring device is

incomprehensible for a man, because it is

represented by a HEX format packet. The data

needs to be stripped of redundant information like

packet numbering and transformed to a recognizable

state – a graph. This is done during the packet

parsing in PDA/Embedded device, where HEX

information 2 bytes in length are transformed to a

binary state, where the data is carried on the first 12

bits. It is transformed to a decimal state and sent for

further processing – sending the data to the server,

storing it in a local database or visualizing it.

The visualization using PDA/Embedded is just a

simple visualization of a curve in real-time or

historical graph. In the server application, the

visualization is far more complex, with the

possibility of storing the current curve as an image,

printing, or zooming.

Neural net of the SOM type on the server is

10x10 in size. The initial weight of each neuron is

random. The weights are assigned progressively by

learning. Finally, the whole structure is ready to

accept data to analyze. Each patient has his own

neural net stored on a server in XML format.

4 CONCLUSIONS

The evolution of Telemedicine is unstoppable and

apparent; therefore ways need to be found to

improve the quality of hospital services, spa services

or hazardous environment workplaces. The area of

software products working in embedded devices in

hazardous environments or PDAs in personal

healthcare is still open and unoccupied. Personal

healthcare products are freely available

(

http://www.omron.com, http://www.weinmann.de) and

with minor hardware modifications they can be used

for data acquisition using wireless Bluetooth, WiFi

or ZigBee technologies. By means of data transfers,

the acquired data can be gathered in database

systems providing access to your personal doctor,

who can be:

• alerted in case of trouble.

• send information and system message to our

application.

• remotely change properties of devices

(change alarm of critical limits in ECG,

e.g.).

Why are there systems available to protect our

property and not our health? Is it not the most

valuable property of ours?

REFERENCES

Janckulík, D., 2007, bachelor thesis, Personal telemetric

system – Personal mobile assistant.

Vašíček, P., 2007, diploma thesis, Personal biotelemetric

system.

Corbelt, BlueEKG datasheet, 2007

http://www.corscience.de; http://www.microsoft.com

http://www.omron.com; http://www.weinmann.de.

Neuron networks prognosis -http://neuron.felk.cvut.cz/

courseware/data/chapter/36nan060/s23.html - (16.4.

2007).

Selforganizing neuron network - SOM, Kohonen maps -

http://automatizace.hw.cz/rservice.php?akce=tisk&cis

loclanku=2006051401 - (13.3.2007).

PERSONAL TELEMETRIC SYSTEM – GUARDIAN

173