Implementation of Contextual e-Healthcare System:

A Prospective e-Service Supported by Context Aware Conceptual

Framework and Image Processing Models

Arif Mahmud, Faria Hossain, Mehadi Hasan and Md. Enjamamul Haque Sarkar

Department of CSE, Daffodil International University, Dhaka, Bangladesh

Keywords: E-healthcare, image processing, IOT, ubiquitous computing.

Abstract: A novel contextual e-healthcare model is implemented based on earlier developed context aware conceptual

framework that systematizes IoT infrastructure to obtain e-services based on context information leaving the

current infrastructure unaffected. This suggested model merges the idea of image filtering models with

communication system where firstly, a system will request and receive the service, a frontend to communicate

with cloud service and lastly, convert the image filter equation into Simulink model. This model has created

Simulink models to convert the mathematical equations of edge detection and FIFO which will reduce analysis

time and user does not need to know any of programming languages. This system has digitized the digital

image processing analysis system through web platform and responsive to all devices such as laptop, tab, cell

phone, computer, etc. Proper health care is one of the elementary rights and the suggested framework model

confirms the availability of services anywhere, anytime without being encircled by any boundary.

1 INTRODUCTION

The internet of things (IoT) can possibly be bounded

in entire network structure based on unvarying and

functional network protocols in which sensible and

practical “objects” are assimilated in the

communication network. „Things‟ can be defined as

a physical object which is capable to communicate

with each other and contribute to the development

of the idea of e-services supported by context

information gained from internet of things (Delphine

Christin, et.al); The perception of IoT immensely

fortifies the e-services especially the e-healthcare.

Formation of a widespread IoT framework can help

to establish ambient computing and ubiquitous

intelligence through internetworking and sharing of

resources among physical entities in dynamic and

configurable networks (Ovidiu Vermesan, et.al).

E-health explains the healthcare exercise based on

the shared application of communication technology

and electronic information. The goal of e-health care

is to advance the medical practice, healthcare

improvement, global networking along with

educational and research work away from the

geographical boundary (Avinandan Mukherjee,

et.al).

The suggested model is formed supported by the

previously formed context aware conceptual IOT

framework

(Theo Kanter, et.al, 2013) and ICTization

framework (Arif Mahmud et.al, 2012) that prolongs

the application of ICT infrastructure. This model will

help to deliver a system to classify detailed collection

of ICT infrastructural features to achieve a precise

strategic placement from the business viewpoint. On

the contrary, this concept will offer a collective

framework with a goal to attain better context aware

pervasive health care services (Shyamal Patel, et.al,

2012).

This method offers e-healthcare facilities by

means of the current network structure along with

modernization of image processing through web

platform. The core infrastructure assists patients with

filtering any image and performs as a communication

media between them and analyzer. This new system

Hossain, F., Mahmud, A., Mehadi, H. and Enjamamul Haque Sarkar, M.

Implementation of Contextual e-Healthcare System: A Prospective e-Service Supported by Context Aware Conceptual Framework and Image Processing Models.

DOI: 10.5220/0009772401490155

In Proceedings of the 1st International Conference of Computer Science and Renewable Energies (ICCSRE 2018), pages 149-155

ISBN: 978-989-758-431-2

149

can request and receive the services from cloud and

can be responsive to all the internet connected

devices, for example, cell phone, tab, computer etc.

As a prolongation of earlier works (Arif Mahmud,

et.al, 2016), (Arif Mahmud et.al, 2012), (Arif

Mahmud, et.al, 2014), (Theo Kanter, et.al, 2013), we

have pointed out some research objectives identified

below and can be simplified through our projected

perceptions explained in the rest of the sections -

• The structure is needed to be supported

through the mutual utilization of communication

technology and electronic information.

• This system is required to be developed for the

sake of educational and research activities.

• This method should have user friendly

features and can filter images without prior

knowledge in computer programming languages.

• The problem solving processes are needed to

be faster and digitalized.

• The system is needed to be based on

accessibility; users should have freedom to access and

to use the services without any interruption.

• The roles of different entities or participants

should be defined and divided in the proposed system

Probable application comprises constant health

observing, disease syndrome investigation, regular

health assessment, elderly monitoring, chronic

disease observation etc (Shyamal Patel, et.al, 2012).

This paper is structured in the following way:

section II explains the context aware conceptual IOT

model; Section III defines the workflow of the

proposed model; Section IV illustrates the

implementation of our proposed context aware

ubiquitous health care model; Section V demonstrates

the prospect of this model and discussions are

provided in section VI.

2 CONTEXT AWARE

FRAMEWORK MODEL

The primary goal of this model is to investigate and

define the smart activities of these smart devices

through maintaining a dynamic communication

among these devices. The suggested framework will

support to systematize IoT infrastructure in order to

obtain e-services based on context information where

the current infrastructure will remain unchanged. The

active association among these heterogeneous

protocols and devices can help to achieve

forthcoming ambient computing in which the utmost

exploitation of cloud computing will be confirmed.

We have divided the total framework system into

4 layers to obtain context aware e-services out of raw

data received from the internet of things as seen in

fig.1. These 4 layers set up a universal framework that

does not revise the existing network infrastructure but

generate an interface among services and objects by

means of network virtualization.

2.1 Connectivity Layer

This layer comprises all the physical objects

involved in the framework model and the

communication among them. Forthcoming internet

mostly depends on the integration of these objects

found in our surroundings and these devices will be

specifically identifiable and manageable. This layer

includes assigning of short range communication

devices such as RFID tags, sensors, actuators, etc. and

resource management verifies the accessibility of

physical resources of these devices and networks

participated in the basic infrastructure. These devices

include very inadequate resources and resource

management confirms the full utilization with slight

overhead. It also permits distribution and sharing of

information among many networks or in a single

network separated into various domains.

2.2 Access Layer

Fig1: Context aware conceptual IoT framework

ICCSRE 2018 - International Conference of Computer Science and Renewable Energies

150

Context data will reach to internet thru IoT

gateway after capturing data by small range

communication devices as raw data. Access layer

contains topology classification, network origination,

formation of network domains etc. This layer also

comprises connection establishment, inter domain

interaction, and transfer scheduling, data

transmissions between sensors and IoT gateway.

Feature management comprises the feature_filter

that receive only accepted context data and additional

data will be rejected. Huge number of sensors

maintains numerous amounts of features whereas

only a trivial subset of features is useful in generating

a context data.

Feature filter aids to lessen irrelevant data

transfer, escalations the data transmission and

diminish energy and unnecessary CPU utilization too.

The number of features can be diverse based on the

type of applications and context data.

2.3 Abstraction Layer

One of the most significant characteristics of

OpenFlow is the addition of virtual layers with the

present layers in which the established infrastructure

will be left unchanged. The system can be considered

as a fully a centralized system from physical layer

standpoint whereas the sharing of services (flow visor

can be used) can be maintained. A central system can

observe, regulate all type of data traffics. It can assist

to attain better reliability, band-width and routing

which will comprise a improved quality of services.

Data are transmitted thru some neighbouring

nodes in a multi-hopping state. Consequently, nodes

which are close to access points accepts more load in

comparable to distant nodes in a downstream

consequence and due to inactivity of the vital nodes,

the network might get collapsed. Virtual presence of

these nodes can resolve the problem where virtual

links can be created among networks through the

negotiation of access points.

2.4 Service Layer

Storage management carries the concept about all

kinds of unacquainted and imperative technologies

that can make the system to be scalable and effective.

This layer is not only responsible for data storage but

also to ensure security. It also permits accessing data

effectually; integrating data to augment service

intelligences, and increases the storage effectiveness.

All business models can receive benefits out of

cloud computing structure for instance, cost and

flexibility of small business perspective and complete

IT problems can be resolved for large organizations.

It will enhance the support for companies, consumers,

employees, distributor etc.

Service management links the needed services

with administrative solutions and thus the new

generation user services becomes simplified. These

upcoming services are required to be unified and

combined to meet the requirements of socio-

economic challenges like environment scrutiny,

security measurement, weather analysis, agriculture

upgrading etc.

3 CONTEXT AWARE

FRAMEWORK MODEL

The proposed model facilities ambient healthcare

services based on previously proposed framework

models. Five entities are included in the system as

seen in figure 2 named as patient, merchants, home

network, network operators and health care centers.

The functionalities of these entities are provided in

brief:

 Patients can utilize the home network to request

and obtain the services and reply to third party

for authentication purposes and desired

services.

 The home network can be formed as a

combination of smart devices such as laptop,

cell phone, PDA etc. with internet connectivity.

 The Merchants works as a middle person to

define and provide the desired services as asked

by the patients. It will filter the images and send

the results to health care centers. It can also

communicate with banks to complete financial

transactions.

 The network operators provide the network

connectivity to handheld devices and govern

the subscriber‟s identity.

 The health care service centers can be

considered as the collection of hospitals,

diagnostics centers along with specialized

doctors, nurses and researchers. The centers

can be distributed in different places and can

share the information databases as required.

Implementation of Contextual e-Healthcare System: A Prospective e-Service Supported by Context Aware Conceptual Framework and

Image Processing Models

151

Fig 2: Proposed healthcare model workflow

 For instance, a patient can demand the service

through the home network both manually or

automatically. For automatic service, either

sensor are used to capture the data for example

patients heart bit, pulse rate, motion etc. or

cameras can capture images intermittently and

send the results to home network. On the

contrary, patient can use the home network to

request services manually.

 Home network guides the information to the

merchant to verify the legitimacy of the user.

The flow of data can be categorized in two

types, such as mobile traffic and IP traffic as

chosen by the patient. Consequently, the user

can appeal the service both via mobile phones

and emails.

 The procedure of sending and receiving

services happened between user and provider

in which the merchant plays the part of the

merchant who becomes connected with both

parties and approves authentication. The

merchant has the vital role as explained in the

previous paper. It can interact with banks to

meet financial transactions. As for examples,

when a subscriber needs to pay for any services

and registrations, merchant will inform the

bank to execute the transaction. At the

completion of financial transaction by the bank,

user and provider will be notified.

 In short, the healthcare centers maintain an

interaction with merchant and network

operators in order to confirm the validity of end

user. After that the demanded services will be

received by the end user via the third party.

4 IMPLEMENTATION AND

RESULT

The following technologies and tools were used

in proposed system:

• Apache server was used as web server

• PHP and CSS were used as web language and

can be applied to HTML and XML.

• MATLAB, Simulink and Xilinx ISE were

used for image analysis

• HTML5, Bootstrap were used for creating the

interface and presenting the content in Web.

Fig 3: Administrative interfaces

This model provides administrative interface for the

system administrator as shown in figure 3. some of

important features of administrative application are

Dashboard log in-out, Profile management using

database, User management, image analysis

(addition, view, examine, solution etc.), result display

etc.

This model also provides a user interface which is

intended for the system user with limited features can

be seen in figure 4. Users can use those services after

getting registered with username, email and

password. Users can provide details information of

the image with addition of images and can also

receive the results.

We have used two filtering methodology, Edge

detection and FIFO and converted their image

processing equation into simulink model as shown in

figure 6 and 5. This project creates block diagram for

the image filter to replace the mathematical equation.

ICCSRE 2018 - International Conference of Computer Science and Renewable Energies

152

These models will be helpful for the third party to

analysis any image without any prior knowledge on

computer programming. Therefore, the processing

time will be faster and becomes convenient to

generate results.

Edge detection refers to the process of identifying

and locating sharp discontinuities in an image. Hence,

it is a vital step in image analysis and it is the key of

solving many complex most image processing

applications to obtain information from the frames as

a precursor step to feature extraction and object

segmentation. It has been used for object recognition,

target tracking, segmentation, data compression, and

image reconstruction.

Our model supports all edge detections techniques

such as Sobel, Prewitt, Roberts and Canny. The

functionalities of these blocks are explained below:

Fig4: User interface

• Image from file: This block takes input from

computer in form of jpeg, png for filtering.

• Color space conversion: This block helps to

convert the true color image RGB to the gray scale

intensity image. This block converts RGB images to

gray scale by eliminating the hue and saturation while

retaining the luminance.

• Edge detection: This block helps to detect the

boundaries of image using Sobel, Prewitt, Roberts

and Canny method.

• Input: This block convert‟s true color image

RGB to the gray scale intensity image.

• Processed image: This block helps to find the

final result of filtered image.

On the other hand, FIFO is used to buffer

temporarily the pixels data for later usage. The FIFO

size is proportional to the length of filters and input

data width. With this method, image will be slightly

blurred. The primary effect of blur image is to reduce

contrast, noise and also to increase visibility of small

object or in detail.

In addition, this method helps to find the actual

shape of any object out of any unclear image. The

functionalities of are given in brief.

• Signal form workspace: Output signal samples

are obtained from the MATLAB workspace at

successive sample times. A signal matrix is

interpreted as having one channel per column. Signal

columns may be buffered into frames by specifying a

number of samples per frame greater than 1.

• An M x N x P signal array creates M x N

matrices at successive sample times. The samples per

frame must be equal to 1 for three-dimensional signal

arrays.

Fig 5: FIFO Simulink model

• Gateway In: Converts Simulink integer,

single, double and fix point to Xilinx fix point or

floating point data type.

• To FIFO: First-in-first out (FIFO) block writes

FIFO data to shared memory storage.

• From FIFO: First-in-first out (FIFO) block

that reads FIFO data to shared memory storage.

• Vertex 2 5 Line Buffer: The block buffers a

sequential stream of pixels to construct 5 lines of

output. Each line is delayed by N samples, where N

is the length of the line. Line 1 is delayed 4*N

samples, each of the following lines are delay by N

fewer samples, and line 5 is a copy of the input.

• 5*5 Filter: The Xilinx 5x5 Filter reference

block is implemented using 5 n-tap MAC FIR Filters.

Nine different 2-D filters have been provided to filter

gray scale images.

Implementation of Contextual e-Healthcare System: A Prospective e-Service Supported by Context Aware Conceptual Framework and

Image Processing Models

153

Fig 6: Edge detection Simulink model

A new context aware real time e-service based on

present network service system has been proposed in

our paper. This system merges the idea of image

filtering with communication system. On the way

three tasks have been finished. Firstly, a system will

request and receive the service, a frontend to

communicate with cloud service and lastly, convert

the image filter equation into Simulink model.

Therefore, all these tasks has been checked and

analyzed for the system. For converting image filter

equation into Simulink model worked with two

methodologies, Edge detection and FIFO. This

system digitized the digital image processing analysis

system through web platform, which is directly

connected to web. Using this system users can easily

analysis the image without any interruption. It will be

faster because it will reduce analysis time and user

does not need

to know any kind of programming language. This

system will be digitized by the combined use of

electronic information and communication

technology. A user can easily upload their image

problem into system and get result from this system.

This system is responsive to all devices like laptop,

tab, cell phone, computer, etc. System can ensure

validity of user can be the same or different person

based on functionality.

Some of the significant solutions as expected from

this model are provided below:

The proposed system is capable of accomplishing

the approval level through maintaining user friendly

features and vital support and services for registration

and electronic payment system.

The system supports accessibility; subscriber has

the liberty to access and consume the facilities.

Importantly the system will not be surrounded by any

borderline.

The system maintains reusability which is a

significant part to incorporate the healthcare service

system with present system. User has the liberty to

choose the processing partner for example, bank,

third party, network operators as required.

Users have access to m-commerce applications

through their hand held devices in real time without

the support of any external devices which allows

users to verify the service validity.

Third party assessments the customer

identification before delivering any services that turns

the system reliable.

The user can carry the hand held device and obtain

the service anywhere within a network coverage

arena that verifies the concept of mobility.

5 DISCUSSION

Our goal is to develop a real time e-healthcare

service system based on present network system. In

this system, we have used several tools and

techniques that make it efficient one and can be

outlined with five entities as proposed in our previous

paper, development of Simulink blocks to convert the

mathematical equations of image processing and a

frontend to communicate with cloud service and to

make user friendly. This model is efficient in logical

partition of physical objects placement, formation of

virtual links among distinct, network domains and

collaborate in multiple applications without the

support of any central management system.

Fig 7: Input and output using edge detection mode

Fig 8: Input and output using FIFO model

ICCSRE 2018 - International Conference of Computer Science and Renewable Energies

154

Context awareness plays a remarkable role in

accomplishing e-services and ubiquitous computing

as well as it permits interpreting of various contexts

received from environments. The obvious IoT

segmentation and certain standard permit various

manufacturers and system vendors to collaborate the

activities and large scale expansion to be fully

operational.

Healthcare services are one of the most

substantial concerns in human life. The utilization of

computerized tools and information increases along

with our demands. However, these improvements are

not sufficient to support new technologies for the

sophisticated machines in modern health care

services. As a consequence, the proposed model will

contribute to the development of secure and

trustworthy system.

To conclude, we have projected a novel

contextual e-healthcare system as an example of the

e-services and it confirms accomplishing the

requirements, purposes and issues being maintained

by the earlier developed framework models. Further,

it is expected that the model will play a noteworthy

role in e-commerce which will become a huge

success in terms of upgrading of e-health care system.

REFERENCES

Delphine Christin, Andreas Reinhardt, Parag S. Mogre and

Ralf St einmet z, “Wireless Sensor Networks and the

Internet of Things: Selected Challenges”, Technische

Universitat Darmstadt, Darmstadt, Germany.

Ovidiu Vermesan, Peter Friess, Patrick Guillemin, Sergio

Gusmeroli, Harald Sundmaeker, Dr. Alessandro

Bassi,Ignacio Soler Jubert, Dr. Margaretha Mazura, Dr.

Mark Harrison,Dr. Markus Eisenhauer, Dr. Pat Doody.

“Internet of Things Strategic, Research Roadmap”.

Avinandan Mukherjee and John McGinnis, “E-healthcare:

an analysis of key themes in research” Montclair State

University, Montclair, New Jersey, USA.

Arif Mahmud and Abdus Sattar, “Deployment of

Contextual E-healthcare System: A prospective e-

service based on context aware conceptual framework

and ICTization framework model”, “11th IEEE

Conference on Industrial Electronics and Applications,

2016,China

Arif Mahmud and Abdus Sattar, “ICTization framework‟:

A conceptual development model through ICT

modernization in Bangladesh”, Dept. of Computer

Science and Engineering, Britannia University, “2nd

International Conference on Advanced Computer

Science Applications and Technologies – IEEE 2013”,

Sarawak, Malaysia.

Arif Mahmud and Abdus Sattar, “Deployment of

Contextual Mobile Payment System: A prospective e-

service based on ICTization framework from

Bangladesh perspective”, Dept. of Computer Science

and Engineering, Britannia University, “International

Conference on Advances in Computer Science and

Electronics Engineering, March 2014”, Kuala Lumpur,

Malaysia.

Theo Kanter, Rahim Rahmani and Arif Mahmud,

“Conceptual Framework for Internet of Things‟

Virtualization via Open Flow in Context-aware

Networks”, Dept. Of Computer and Systems Sciences,

Stockholm University, Sweden, IJCSI International

Journal of Computer Science Issues, Vol. 10, Issue 6,

No 1, November 2013.

Shyamal Patel, Hyung Park, Paolo Bonato, Leighton Chan

and Mary RodgersEmail, "A review of wearable

sensors and systems with application in rehabilitation",

Journal of Neuro Engineering and Rehabilitation 2012.

Keng Siau and Peter B. Southard, “E-healthcare strategies

and implementation”, Department of Management,

College of Business Administration, University of

Nebraska-Lincoln, Lincoln, NE, 2002

Keng Siau and Peter B. Southard, "E-healthcare strategies

and implementation", Department of Accounting and

Information Systems, College of Business

Administration, Texas A&M International University,

5201, University Boulevard, Laredo, TX 78041, USA,

Int. J. Healthcare Technology and Management, Vol. 4,

Enterprises Ltd.

F. D. Davis and V. Venkatesh, "Measuring user acceptance

of emerging information technologies: An assessment

of possible method biases," in Proc. 28th Hawaii Int.

Conf. System Sciences, 1995, pp. 729- 736.

B. N. Schilit and M. M. Theimer, "Disseminating active

map information to mobile hosts," IEEE Network, vol.

8, no. 5, pp. Complete Petri Net Model 22-32, 19.

Nathalie Bricon-Souf and Conrad R. Newman, "Context

medical informatics 76(2007)2–12.

Fu-Shiung Hsieh, “Context-aware Workflow Driven

Resource Allocation for e-Healthcare” Department of

Computer Science and Information Engineering

Chaoyang University of Technology, Taichung County,

Implementation of Contextual e-Healthcare System: A Prospective e-Service Supported by Context Aware Conceptual Framework and

Image Processing Models

155