MultiScience - XXXIII. microCAD International Multidisciplinary Scientific Conference University of Miskolc, 23-24 May, 2019, ISBN 978-963-358-177-3
TELEMEDICINE SYSTEMS
Dániel SZÖLLŐSI
Junior researcher, Smart division, Bay Zoltán Nonprofit Ltd. for Applied Research, Miskolc
ABSTRACT
Thanks to the advanced health care system and high living standards the life expectancy is continuously increasing. As the human mobility also increased, family members often live far from each other. This means that caring for the elders shifts from the younger generation to the social care system. This adds extra burden to the health care system. This may be eased by integrating IT systems designed for elderly users. However, the Telemedicine system is not only limited for social care, it can be used in any home care situations for anyone who needs a continuous supervision, or who lives far from any medical care facility. Utilization of the system shall fit into the daily routine of any of the users and its application should not be a challenge. This newly developed solution can be applicable from hospitals through clinics to nursing houses. This system gives greater freedom and self- confidence to its users. On the other hand, it provides meaningful information to their relatives or the formal caregivers. This way the doctors can get a summary and a definite picture of the patients' condition. Whether the measurements are done by a patient or a caregiver we have to assure that those are made in a right way. This may be achieved by showing the users the usage of devices, so they can gain confidence to use the system. After that the patients are able to do the measurements by themselves at their home with no disturbing factor. All of the measurements – blood pressure, blood glucose, body temperature and balance scale – can be personalized when and how the measurement should be done and what the measurement limits are.
INTRODUCTION
The big question is: how the supervisor caregivers will be able to check their patients’ condition if they are at home and all the measurements are done by them.
The Telemedicine system has a very helpful website with which the above- mentioned problem can be solved. The caregivers are able to see their patients’
measurements anytime, anywhere, since we provide a 7/24 service.
A user (patient, doctor or nurse) is able to query his/her measurements or a doctor or nurse its patients’ measurements by date and type. Furthermore, caregivers can see some of the patients’ data, the assigned devices, and are able to modify the measurement limits individually. The patients are only able to query their own measurements. The measurement results are displayed in different charts
DOI: 10.26649/musci.2019.028
in the web client. However, the data that are displayed on the website should come from somewhere. Now four devices are integrated in the system:
Blood pressure meter;
Blood glucose meter;
Body thermometer;
Balance scale;
Neither of them is able to directly communicate with the web server. Since they only have Bluetooth module or Wi-Fi module (the Balance Scale). So, we needed a communication device, a phone – acting as a communication HUB - and an application that is able to communicate with the devices and with the web server.
The phone receives the data from the devices and adds some necessary information to it, like date, time, measurement place and the supervisor. Or if we are talking about the Balance Scale it builds an XML document from the incoming data.
The documents that are sent to the web server need to be processed and saved to a database. For this there is a service where the HUB sends the data and this service is also responsible for the processing. The processed data is saved to the database in XML format. Furthermore, the application can identify any registered user. Every user must have an NFC tag that identifies them. The NFC tags value is saved in the database and the underlying user is queried by that by the HUB. This way the system can assign the measurements to a patient and a supervisor for him/her.
METHOD OF MEASURING Monitoring at home
Vital parameters: blood pressure, blood glucose levels, body temperature, body weight and balance. These are monitored by the patients at their home by the doctor’s advice. Within the system we have created tools that send warning messages to doctors or the nurses if there are any measurements over or under the limits. [1]
HUB
A smart phone with our application is responsible for transmitting information from the terminal equipment (blood pressure meter, blood glucose meter, body thermometer and Balance Scale) to the Telemedicine web service and identifying users by scanning their NFC (Near Field Communication) tags. The HUB also builds an XML document from the information that comes from the Balance Scale.
Furthermore, the HUB displays the received information.
On the other hand, a user is able to pair new devices with the application;
therefore, there is a built-in guide to help the users with that. Also, there is a hidden
Admin mode in the HUB where we can set the webserver address, weather we would like to use mobile internet or Wi-Fi with SSID and password, the login timeout –after what time the logged in user should be logged out- and if we want to remove the measured vales after a while.
Fig. 1
Blood pressure, blood glucose and body temperature measurement
It is easy to use these devices, only a bit of attention is required. These equipment’s communicate with the HUB via Bluetooth (Low Energy). They generate a JSON document after every measurement and are sent to the phone.
Fig. 2
Balance Scale measurement
Balance Scale should be taken as a regular bathroom scale. The difference is the measurement's time and the position in which the measurement should be done.
In this case these should be done with:
closed eyes and outstretched arms,
closed eyes and arms clamped next to the body or
open eyes and arms clamped next to the body
During a measurement the device not only measures the user's weight but it gathers data of its movement. This takes about 30 seconds. The measurement has 4 parts:
the weight measurement and
the balance measurement broken into 3 parts.
Every section is indicated by a beep noise by the Balance Scale, just like the end of the measurement.
Because in the first ten seconds the patient is trying to stabilize him/herself. In the next ten second, we can say the patient is standing as still as he/she can. If the patient is tired or has some other problems in the third ten seconds, he/she will start moving again or more.
Balance Scale communication
This is the only device that uses Wi-Fi for communication with the HUB.
The main problem with that is how these devices find each other; they must be on the same network and must know each other’s IP addresses. This is really hard to find if the WLAN network uses DHCP. We implemented an IP discovering method which sends a pocket to the connected networks broadcast address. If our HUB is connected to that network it will answer for that. From this communication both the Hub and the Balance Scale will can extract each other’s IP addresses.
After they found each other the Balance Scale, goes into sleep mode. For receiving information from the Balance Scale we use a Socket server in the application which reads the data line by line, since it is coming as a time series of data. Each line starts with a specific code that determines what kind of data it is or whether all the data is sent; the information that comes from the pressure gauge stamps are not marked with any code, they are sent separated by a coma. The reason why the Balance Scale sends its data like that is that it is built with a basic ESP32 chipset, which is not capable of building complex XML documents. Therefore the document is built inside the HUBs Socket receiver. Another problem with the Wi-Fi is that it is not so reliable. Some cases the connection goes off which must be
handled on both sides; on the Balance Scale the sending should be stopped and should be retried, on the HUB the Socket server and the underlying XML builder should be released and restarted.
If everything went well and the XML document is built successfully the HUB tries to send it to the web server, but whether it was successful or not, the document is saved to the phone's internal storage. That applies to the other measurements as well. Only the last 10 measurements are saved on the phone, the least recent ones are always deleted.
PROCESSING INCOMING DATA
Processing in the HUB
For blood pressure meter, blood glucose meter and thermometer the HUB does not process their data because they are sending a whole JSON document. The measurement time is modified, a message property, patient identifier, the supervisor’s identifier –if there is not the property is omitted- and for the thermometer the place of the measurement is added.
However, for the Balance Scale we have to process all the incoming data.
Since it is sent as lines of a plain text. As mentioned above the Balance Scale sends it each line beginning with a code so we can sort those out and put it in the XML document. Yet the data coming from the pressure gauge stamps are not marked.
Building an XML document on any framework, device or with any language is very precious and resource demanding even on a phone.
The system (Android OS and JAVA) does not help us by easing out our way to build these documents. Because it has a very basic XML document builder where you have to manually open and close tags, add attributes to them or write texts into them. Therefore, our builder logic is grown quite big and complex, just as the documents that are built with it.
Communication between the HUB and the server
Depending on the environment the Hub is able to send the measurement documents via mobile internet or Wi-Fi. The documents are sent to a web service.
Processing in the web service
This web service is responsible for creating usable data for the front end and saving the raw and the processed measurement in the database. Every processed measurement is saved in an XML document.
From the JSON documents the measured value, the place of the measurement is saved in the XML. The other properties of the JSON are parsed and saved into the database columns like the time, message, patient identifier and the supervisor identifier, these are the necessary information’s that we need to identify a measurement. So basically, it is a simple conversion from JSON to XML.
The Balance Scale measurement comes in XML format and contains a few thousands of data. The necessary information’s are extracted from the document as above mentioned. It is a resource and time consuming-operation to parse and process the pressure gauge values.
We calculate points from these values as shown below:
Fig. 3 Dimensions of the Balance Scale
Alpha is the angle of the diagonals of the rectangle, the dimensions of the Balance Scale can be extracted from the original XML document. Fx is the pressure gauge stamp.
From the calculated points we are able to calculate the sway length and the stability index.
Sway length (SL) [2]:
AP(Y) anteroposterior (from front to back)
ML(X): mediolateral (side to side)
DISPLAYING MEASUREMENTS
We created the web site in collaboration with the caregivers. This way it only contains the most useful and necessary information’s. We used a popular chart library on the web site the so called Chart.js.
Blood pressure, blood glucose and body temperature measurements
We used line charts to display each queried measurement. With that everyone can see the difference between the measurements.
Fig. 4 Blood pressure display
Balance Scale measurements
The measurements are displayed in a line chart also. But since it contains more complex information’s we added a bar, a linear and a modified line chart.
In the bar chart we display the side loading.
The linear graph is responsible for displaying the patients’ movement. The doctors can read off where the patient’s weight center is located; it is divided into three parts as mentioned earlier.
Fig. 4 Balance Scale overview display
In the last diagram we display how the much the patient moved between each sampling.
Fig. 5 Balance Scale sway display
CONCLUSION
We deployed a test system from we gathered a lot important information’s with dozens of involved test users.
From the caregivers’ side we assessed that the system is capable of help the caregiver’s job by giving meaningful information’s and accelerate their work. From the patients’ side we have got a different conclusion. Because the newly introduced system attracts large number of people. But the users’ willingness to use the system is decreases in time. This is proved by the feed backs of the caregivers and the system usage diagnostics.
We are also planning to include more functionality in the system. We will incorporate smartwatches for heart rate and activity measurement and for indoor localization and wandering detection by porting our indoor localization system [4]
to a smartwatch.
However, seeing the whole picture, we can get the conclusion of the system has a bright future in the health care. Until then we have work on it as our test users suggests improvements or finds bugs here and there in the system.
REFERNCES
[1] L. Árvai, D. Perényi and D. Vass, “Organisational life assistant: How an IT system can help for the ageing society,” 2018 19th International Carpathian Control Conference (ICCC), Szilvasvarad, 2018, pp. 395-399.,
DOI: 10.1109/CarpathianCC.2018.8399662
[2] „Measures of postural fluctuations,” [Online]. Available:
http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0113897., DOI:
10.1371/journal.pone.0113897
[3] „Effectiveness of elastic band-type ankle–foot orthoses on postural control in poststroke elderly patients as determined using combined measurement of the stability index and body weight-bearing ratio,” [Online]. Available:
https://www.dovepress.com/effectiveness-of-elastic-band-type-anklendashfoot- orthoses-on-postural-peer-reviewed-fulltext-article-CIA., DOI:
10.2147/CIA.S92888
[4] L. Árvai, "Mobile phone based indoor navigation system for blind and visually impaired people: VUK — Visionless supporting frameworK" 2018 19th International Carpathian Control Conference (ICCC), Szilvasvarad, pp.
383-388., DOI: 10.1109/CarpathianCC.2018.8399660
