Embedded Internet for Pulse Oximeters I Embedded Internet for Pulse Oximeters I by Matthew Roy Burey School of Information Technology and Electrical Engineering University of Queensland Submitted for the degree of Bachelor of Engineering in the division of Computer Systems October 2001
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Embedded Internet for Pulse Oximeters I
Embedded Internet for
Pulse Oximeters I
by
Matthew Roy Burey
School of Information Technology and Electrical Engineering
University of Queensland
Submitted for the degree of
Bachelor of Engineering
in the division of Computer Systems
October 2001
Embedded Internet for Pulse Oximeters I
King’s College
Upland Rd
St Lucia Qld 4067
19 October 2001
Head of School
School of ITEE
The University of Queensland
St Lucia QLD 4067
Dear Professor Kaplan,
In accordance with the requirements of the degree of Bachelor of Engineering in the
division of Computer Systems Engineering, I present this thesis titled “Embedded
Internet for Pulse Oximeters I”. This work was performed under the supervision of
Dr Stephen Wilson.
I declare that the work submitted in this thesis is my own, except as acknowledged in
the text and footnotes, and that it has not previously been submitted for a degree at the
University of Queensland or any other institution.
Yours Sincerely
Matthew Burey
Embedded Internet for Pulse Oximeters I
i
Acknowledgements I would like to thank Steve Wilson for his guidance and supervision during this year.
Special mention goes to the hard working people at the Mater Children’s Hospital
especially Gordon Williams who helped with this project. Last but not least I would
like to thank Ben Lever, my thesis partner, for his hard work and co-operation whilst
implementing the client side.
Thank you.
Embedded Internet for Pulse Oximeters I
ii
Abstract
In recent years there has been many investigations into sleeping disorders. Many of
these studies are carried out in specially equipped units in which a patient is
monitored whilst sleeping. Measurements that are taken are in the form of
2.0 LITERATURE REVIEW................................................................................6 2.1 TELEMEDICINE - REMOTE MEDICINE ................................................................... 6
2.1.2 Hospital Without Walls ................................................................................ 6 2.1.3 CARE System................................................................................................ 7
2.2 EMBEDDED INTERNET TECHNOLOGY ................................................................... 7 2.2.1 Embedded Internet Possibilities................................................................... 8 2.2.2 Embedded Internet In Practice .................................................................... 9 2.2.3 Available Embedded Web Servers.............................................................. 10
2.2.3.1 Netburner............................................................................................. 10 2.2.3.2 Picoweb ............................................................................................... 11 2.2.3.3 Connect One iChip .............................................................................. 12 2.2.3.4 Seiko S7600A...................................................................................... 12 2.2.3.5 Rabbit 2000 TCP/IP Development Kit................................................ 13 2.2.3.6 Review of Available Web Servers ...................................................... 14
2.3 PULSE OXIMETER TECHNOLOGY ........................................................................ 15 2.3.1 Oxinet II Central Station Network ............................................................. 15 2.3.2 TeleOximetry – 3900/3900P Pulse Oximeter............................................. 16 2.3.3 CSI-TC Wireless Monitoring Station ......................................................... 16 2.3.4 Internet Polysomnography ......................................................................... 17
3.2.1 Real time Data Acquisition ........................................................................ 21 3.2.2 Trend Data ................................................................................................. 24 3.2.3 General Status Information........................................................................ 26
3.3 FINAL PRODUCT ................................................................................................. 28 3.3.1 Displays...................................................................................................... 29 3.3.2 In Practice .................................................................................................. 31
4.1.1 Real time Data Acquisition ........................................................................ 33 4.1.2 Trend Data ................................................................................................. 34 4.1.3 General Status Information........................................................................ 34
4.2 REVIEW OF AVAILABLE SYSTEMS ...................................................................... 35 4.3 EFFECTIVENESS OF DESIGN ................................................................................ 36 4.4 COMMERCIAL VIABILITY ................................................................................... 37
5.0 FUTURE WORK ...........................................................................................38 5.1 SUPPORT FOR OTHER DEVICES ............................................................................ 38 5.2 DIFFERENT INTERNET CONNECTIONS ................................................................. 39 5.3 SECURITY........................................................................................................... 40
v
5.4 REMOTE FIRMWARE UPDATES ........................................................................... 41 5.5 MULTI-USER MULTI CHANNEL SYSTEM ............................................................. 41
8.0 APPENDIX A .................................................................................................46
9.0 APPENDIX B .................................................................................................55
Embedded Internet for Pulse Oximeters I
vi
LIST OF FIGURES Figure 1-1: Pulse Oximetry Diagram............................................................................ 2 Figure 1-2: Oxipleth Pulse Oximeter from Novametrix. Photo courtesy of
Novametrix............................................................................................................ 2 Figure 1-3: Mater Children’s Sleep Unit ...................................................................... 3 Figure 1-4: System overview. Diagram courtesy of Ben Lever................................... 4 Figure 1-5: Implementation breakdown........................................................................ 5 Figure 2-1: Simplified block diagram of "Hospital without walls" implementation in
a single home. Diagram courtesy of The Center for Online Health..................... 7 Figure 2-2: Internet Fridge. Picture courtesy of LG..................................................... 9 Figure 2-3: Netburner CFV2-40 - Courtesy of Netburner. ......................................... 10 Figure 2-4: Picoweb server courtesy of Lightner Engineering ................................... 11 Figure 2-5: iChip block diagram. Picture is courtesy of Connect One. .................... 12 Figure 2-6: Seiko S7600A Block Diagram. Picture courtesy of Seiko Instruments. . 13 Figure 2-7: Rabbit 2000 TCP/IP Development Kit. Picture courtesy of Rabbit
Semiconductor..................................................................................................... 14 Figure 2-8: Oxinet II Central Station Network. Picture courtesy of Nellcor. ............ 15 Figure 2-9: 3900/3900P Pulse Oximeter. Picture courtesy of Datex-Ohmeda. ......... 16 Figure 2-10: Wireless Monitoring Station. Photo is courtesy of Criticare................. 17 Figure 2-11: Internet Polysomnography device. Courtesy of Advanced Medical
Electronics (AME). ............................................................................................. 18 Figure 2-12: Positioned on patient. Courtesy of AME. ............................................. 18 Figure 3-1: System overview ...................................................................................... 19 Figure 3-2: File interaction diagram............................................................................ 21 Figure 3-3: Real time mode functional overview........................................................ 22 Figure 3-4: Trend mode functional overview ............................................................. 25 Figure 3-5: Patient details functional overview .......................................................... 27 Figure 3-6: Final product (front view) ........................................................................ 28 Figure 3-7: Final product (back panel)........................................................................ 28 Figure 3-8: oi2 system operating in the sleep unit ....................................................... 31 Figure 3-9: Staff can easily set up the oi2 system ....................................................... 31 Figure 5-1: Satellite Implementation. Diagram courtesy of Ben Lever. .................... 40 Figure 5-2: Multiple Patient Scenario. Diagram courtesy of Ben Lever. .................. 42 Figure 9-1: Screen shot of main menu page................................................................ 55 Figure 9-2: Screen shot of patient details page ........................................................... 56 Figure 9-3: Screen shot of clear trend page................................................................. 56 Figure 9-4: Screen shot of date and time page ............................................................ 56 Figure 9-5: Screen shot of real time page ................................................................... 57 Figure 9-6: Screen shot of time out page .................................................................... 57 Figure 9-7: Screen shot of trend dump page ............................................................... 58 Figure 9-8: Screen shot of the online help page.......................................................... 58
Embedded Internet for Pulse Oximeters I
vii
LIST OF TABLES
Table 3-1: Real time data format................................................................................. 21 Table 3-2: NOVACOM1 mode 1 data ........................................................................ 22 Table 3-3: Real time mode test data............................................................................ 23 Table 3-4: Trend data format ...................................................................................... 24 Table 3-5: Test trend dump data ................................................................................. 26
Embedded Internet for Pulse Oximeters I
1
1.0 Introduction
Sleeping disorders are a common ailment for young children. As many as 30% [1] of
all children at some stage suffer from some form of sleeping disorder, many of which
will never receive the proper diagnosis or treatment. Ineffective sleep has a great
impact on growth and development and also contributes to poor academic results.
The immune system and general health, both mental and physical, are also affected.
Given that such a large number of children suffer from poor sleep it is not surprising
that research has been undertaken in this field. A large number of disorders have been
defined and categorized into three main categories. These are insomnia, sleep apnea
and narcolepsy, which in layman’s terms are not enough sleep, disturbed sleep and
too much sleep respectively.
Sleep apnea affects the breathing patterns of a sleeping person. A person who suffers
from this does not breath properly throughout the night and they may go through brief
periods when breathing stops all together. There are two main types of sleep apnea,
these are, obstructive sleep apnea and central sleep apnea. Obstructive sleep apnea is
the hindrance of the airway from a partial or total blocking of the throat. In adults this
can be caused by inherent characteristics, excessive obesity, and alcohol consumption
prior to sleeping. In children large tonsils and cranio facial deformities often cause
this. Central sleep apnea is caused by a delayed breathing signal from the brain. In
both cases sleep is broken so that breathing can continue. This may occur hundreds of
times during the night and the patient may not have any recollection of the waking
periods.
Polysomnography (PSG) is the recording of physiological variables that are related to
the stages of sleep in an effort to diagnose causes of sleep disorders [3]. The
physiological variables that are recorded are:
!"Electroencephalography (EEG)
!"Electro-oculography (EOG)
!"Electrocardiography (ECG)
Embedded Internet for Pulse Oximeters I
2
!"Electromyography (EMG)
!"Airflow
!"Respiratory effort
!"Sound recordings to measure snoring
!"Video monitoring of body positions
!"Core body temperature
!"Pulse Oximetry
Pulse oximetry is the measurement of pulse rate and oxygen saturation of the arterial
blood. In essence it works by passing two different wavelengths of light through a
peripheral part of the body such as a finger. This light is absorbed depending on the
oxygen saturation of hemoglobin in the blood cells. As shown in Figure 1-1 a photo-
sensor detects the light on the other side of the peripheral and the resultant oxygen
saturation is calculated. The light absorbed also consists of a component that
corresponds to the pulse rate. This is a result of the increase in arterial blood volume
with each heartbeat. Pulse oximetry is one of the physiological variables that are
measured during the diagnosis of sleeping disorders. It is a useful variable to monitor
in cases such as sleep apnea as the oxygen saturation of the blood directly corresponds
to the volume of intake air.
Figure 1-1: Pulse Oximetry Diagram
Figure 1-2: Oxipleth Pulse Oximeter from Novametrix. Photo courtesy of Novametrix.
Diagnosis of sleeping disorders is undertaken in a controlled and monitored
environment. Hospitals such as the Mater Children’s in Brisbane have constructed a
specially designed unit consisting of four sleep rooms. These rooms are equipped
with probes to monitor vital signs during the different stages of sleep. Monitoring
equipment is located outside the room so they are accessible whilst the child is
Embedded Internet for Pulse Oximeters I
3
sleeping. A data logging system is in place, which is a server running the Uniquant
Sleep Acquisition System. The data is logged throughout the night and two medical
staff also monitors the patient. From the ascertained data an assessment of the
patients stages of sleep can be made, which leads to the diagnosis of disorders.
Figure 1-3: Mater Children’s Sleep Unit
Since the Mater Children’s Hospital Sleep Unit is the only such unit in Queensland
and upper New South Wales the limited number of rooms poses quite a problem.
Given that 3%[1] of children suffer from obstructive sleep apnea at some stage it can
be seen that the resources to cope with such a high demand of cases does not currently
exist. Further more, the associated cost of keeping a child overnight is relatively high.
In a rural situation, a family would have to travel into Brisbane, stay overnight and
return the next day. If the study was carried out over multiple nights then the cost for
the family and hospital rapidly increases. Not only cost but also the family is
inconvenienced as mother and/or father would miss work and possibly a baby-sitter
would need to be arranged for the other children. As a result it is obvious that a better
solution needs to be incorporated.
Remote monitoring of such equipment allows for a more flexible situation. The
expense of sending a small box via a courier to a local hospital or to the patient’s
home is insignificant in comparison to the previous case. The inconvenience is also
reduced tenfold. Remote monitoring also brings with it the possibility of specialist
doctors being able to monitor their patients from a distance.
Embedded Internet for Pulse Oximeters I
4
In building an “embedded internet for pulse oximeter” a door is opened on the
possibility of remote monitoring via the Internet. The proposal is put forward that an
embedded Internet web server device be designed so that remote monitoring can be
enabled on a pulse oximeter. The for-mentioned device, named Oximeter Internet
Interface or oi2, is a web server connected via RS-232 serial interface to a pulse
oximeter. A system overview is shown in Figure 1-4. The oi2 system is able to serve
pulse, SpO2 data and also trend information to the Internet. Medical staff can then
download the trend data or view real time streaming data from any web browser on an
Internet capable device. Patient details can be stored locally on the oi2 system so that
identification is possible. The oi2 system will have enough control so that it can clear
the trend memory located on the pulse oximeter.
Figure 1-4: System overview. Diagram courtesy of Ben Lever.
The design of an Internet based solution can be broken down into two major
components, these being the server side and the client side. This is shown in Figure
1-5. The server side includes the communication with the pulse oximeter and serving
the information to the Internet. The client side includes the retrieval of the information
from the Internet and displaying that information in a human readable form. Basically
the server side implementation executes on the server where the client side is
transferred to the client and executes there. This thesis is mainly concerned with the
server side implementation of the system.
Embedded Internet for Pulse Oximeters I
5
Figure 1-5: Implementation breakdown
This solution needs to be inexpensive relative to the cost of a pulse oximeter and also
robust as it will be portable. Ease of use is another major factor when designing such
a system. Simplifying the operation of the device allows for less technical operators
to adequately use the equipment to it’s full potential.
The Internet serves as a global network for remote monitoring of devices. Since the
Internet infrastructure is already extensively in place the cost of setting up a remote
monitoring facility is greatly reduced. A wealth of resources is available for the
Internet particularly in relation development and standard protocols used. This
reduces cost and time in development of such a system. The Internet is also a
somewhat familiar medium, as it is becoming a part of everyday life for most people.
Embedded Internet for Pulse Oximeters I
6
2.0 Literature Review This section provides an overview of the current technology available and its
relevance to this thesis. This technology relates to the current advances in embedded
Internet, pulse oximeters, and remote sleep medicine.
2.1 Telemedicine - Remote Medicine
Telemedicine utilizes current telecommunications technology such as
videoconferencing to help support those requiring health care. This is common in
situations where people are unable to access conventional health care. These include
but are not limited to the elderly and rural patients. Specialist care to rural areas is
also possible through the means of telemedicine. In the past, telemedicine has been
limited in transmission medium but currently advances are being made to incorporate
today’s technology.
2.1.2 Hospital Without Walls
“Hospital without walls”[13] is a documented project that encompasses the needs of
elderly patients living at home. By using current telecommunications and information
technology practices patients can be moved from hospitals/nursing homes back to
their own home. Patients are monitored with various probes and information is
transmitted back to a base station through a wireless local area network link. At the
base station the information is collated and sent through to an assessment center for
analysis. The base station consists of a PC in which data can also be entered
manually by visiting nurses. As a first implementation the sensing technology used
was to measure heart rate and blood pressure as well as attitude, gait and fall status.
Accelerometers are used to sense a patient falling and generate the appropriate alarm
at the assessment center. This system provides a better quality of life, as there are
definite benefits to living in familiar surroundings. Not only this but the cost savings
for home telemedicine are quite substantial.
Embedded Internet for Pulse Oximeters I
7
Figure 2-1: Simplified block diagram of "Hospital without walls" implementation in a single
home. Diagram courtesy of The Center for Online Health.
2.1.3 CARE System Mastushita Electrics Works (MEW) has developed the CARE system [6] using
emWare’s device internetworking technology. The CARE system is targeted for the
elderly in nursing homes. Conventional methods used a mere camera that was viewed
by an attendant. This could be set up with multiple screens or with one screen
changing between cameras. Both these scenarios have problems as events such as a
patient falling down could be missed. The system comprises a camera and sensors
placed in a room within the nursing home. The sensors can detect locality and
whether the patient is lying down or standing up. Within the CARE monitor the
inputs are polled and events and alarms such as “laid down on bed” and “agitated
sleep” are generated. The events and alarms are transmitted back to central control
station. These are displayed in a web browser style interface and alert the attendant to
any problems. The system is set up over Ethernet but runs a proprietary open network
protocol called emNet. This system is in use in nursing homes in Japan.
2.2 Embedded Internet Technology
Embedded Internet technology is the current trend to manufacture small web servers
and place them in devices to allow control and monitoring via the Internet. Devices
such as printers, routers, cameras, and set-top boxes have embedded Internet devices
Embedded Internet for Pulse Oximeters I
8
built into them so as to allow easy configuration and status information. The Internet
gives a standardized and familiar approach, which is relatively low cost. More
mainstream user interfaces include extra hardware such as keypads and screens,
which are suddenly not needed when implementing embedded Internet interface.
Embedded Internet technology has many other advantages, which allow devices to be
controlled remotely. Hard to reach equipment can now be monitored, maintained and
controlled from a distance and user manuals can be stored on-line for easy access.
2.2.1 Embedded Internet Possibilities
There are many uses of embedded Internet, some of which already exist in
commercially available products. Not only does embedded Internet simplify control,
diagnostics, maintenance and monitoring but also cuts the cost. This cost cutting is
shown in development and in practice as well. Just about anything can benefit from
this added feature, which has been demonstrated by such devices as the Internet
Fridge. From the fridge the user can surf the web, retrieve email, watch TV, play
MP3’s, download recipes and a hosts of others. This is all accomplished by a 15”
LCD touch screen with a virtual keyboard much the same as some PDA’s. This is an
example of what some people would call a novelty more than practical application but
clearly demonstrates that a seemingly unobvious device can benefit from embedded
Internet technology. All this shows that there are many uses to embedded Internet
technology some serious and some not so serious.
Embedded Internet for Pulse Oximeters I
9
Figure 2-2: Internet Fridge. Picture courtesy of LG.
2.2.2 Embedded Internet In Practice
Many commercially available products are having the added feature of embedded
Internet built into them. In some cases the extra expense is warranted, in others
though it is more a novelty than anything else.
A good example of embedded Internet is the HP LaserJet series of printers. This
series has had web servers embedded into them to allow remote printing, monitoring
and a host of other useful features. The embedded Internet changes the way users
access, manage and print information. As a forecast from a market analyst group,
35% [17] of devices attached to the Internet by 2006 will be non-PC based. The
printers in this series come with the added features of HP smart chip. This enables the
network administrator to receive notification by email, mobile phone or computer
about supply of toner or paper. The HP LaserJet 4100 costs about $US1099 [17] and is
the top of the range printer. The HP LaserJet 1200 still has embedded Internet
technology but comes in at a price of $US399.
Another practical application of embedded Internet is the myriad of camera set up on
beaches around the world that have a direct link to a web site. In this way, surfers can
go on the web to view the beach conditions and decide whether it is worthwhile going
Embedded Internet for Pulse Oximeters I
10
surfing. There are quite a few beaches with the installed even here in Australia. A
good example is Noosa in Queensland.
2.2.3 Available Embedded Web Servers
There are quite a few embedded web servers on the market today. These vary greatly
in specifications and price. This section covers some of the embedded web server
technology that was considered when originally researching the topic.
2.2.3.1 Netburner
The Netburner [9] package is Motorola Coldfire processor running the real time
operating system uC/OS. The netburner has a fully implemented TCP/IP stack with
UDP and supports HTTP, PPP, FTP, and TELNET. The NetBurner CFV2-40 starts at
around US $499 [10]. This includes the Motorola Coldfire 5206e, which is a 40 MIPS
processor, which has 512 kBytes of FLASH compressible up to 1 Mbytes. Available
to the processor is 4 Mbytes of DRAM and an 8Kbyte allocation of non-volatile
parameter storage. The CFV2-40 includes 2 DMA channels and dual UARTs. Also
included in the design is a 60-pin interface to attach custom hardware. These kits are
designed to get projects running quickly and easily.
Figure 2-3: Netburner CFV2-40 - Courtesy of Netburner.
Embedded Internet for Pulse Oximeters I
11
2.2.3.2 Picoweb
The Picoweb is a miniature web server from Lightner Engineering. The design of the
Picoweb is based around the ATMEL 8515 micro-controller and also includes 8k
FLASH, 512 bytes of EEPROM, 512 bytes of RAM and 32k serial EEPROM. The
8515 is an 8 MIPS processor. The Realtek 8039 handles the network interface. This
is an NE2000 compatible controller.
Previously the Picoweb was published in some technical magazines as the worlds
smallest web server. The, first prototype, was claimed to be priced at a modest $25
for a breadboard design. Later versions included more serial EEPROM for web site
storage and were packaged as a commercially available product. Lightner
Engineering now sell the full development kits for $149[11].
Figure 2-4: Picoweb server courtesy of Lightner Engineering
The Picoweb includes a real time network kernel, TCP/IP stack and HTTP web
server. Network connectivity is established with an RJ-45 type connection.
Programming is accomplished through the parallel port of a PC and the Picoweb can
interface to other devices through the serial interface or the extra port on the 8515
micro-controller. All these pins have been grouped together on one 25 pin D-shell
connection.
Embedded Internet for Pulse Oximeters I
12
2.2.3.3 Connect One iChip
The iChip [12] is a co-processor that communicates with the existing processor to
facilitate a connection to the Internet. Being an application-specific Internet
controller the iChip can be connected easily to an existing design. The iChip is
compatible with Connect One’s proprietary API, simplifying communication between
the host processor and the iChip. With a fully implemented TCP/IP stack the iChip
can easily be connected to the Internet via several mediums. With the iChip changing
the design between modem and LAN connections becomes an insignificant step.
Figure 2-5: iChip block diagram. Picture is courtesy of Connect One.
2.2.3.4 Seiko S7600A
Seiko have produced a fully self-contained TCP/IP stack in a chip called the S7600A [18]. This LSI device features TCP/IP version 4, UDP and PPP protocols and
incorporates a 68/80 MOTO/Intel microprocessor bus interface. For network
interaction the TCP/IP stack utilizes a 10Kbytes block of SRAM. Low power
consumption and a wide operating voltage help make this an easy to integrate
solution. To help push this product Seiko have a comprehensive listing of the full
technical specifications online that includes development APIs, source code and
utilities. Figure 2-6 shows the block diagram of the device showing the