- Email: [email protected]
Mobile metadata assisted community database of chronic wound images
Accepted Manuscript Title: Mobile Metadata Assisted Community Database of Chronic Wound Images Author: id="aut0005" > Chinmay Chakraborty id="aut0010" > Bharat Gupta id="aut0015" > Soumya K. Ghosh PII: DOI: Reference:
S2213-9095(14)00044-5 http://dx.doi.org/doi:10.1016/j.wndm.2014.09.002 WNDM 45
To appear in: Received date: Accepted date:
11-7-2014 29-9-2014
Please cite this article as: Chakraborty C, Gupta B, Ghosh SK, Mobile Metadata Assisted Community Database of Chronic Wound Images, Wound Med. (2014), http://dx.doi.org/10.1016/j.wndm.2014.09.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Mobile Metadata Assisted Community Database of Chronic Wound Images Chinmay Chakraborty, Bharat Gupta, and Soumya K. Ghosh
an
us
cr
ip t
patient but not focused on large number of patients data handling techniques. Portable devices like networked mobile camera phone can be used to capture the image because of it is a more powerful, low processing power device, handling complex data, presents graphical user interface (GUI) and user friendly in real-time manner at a distance [6] [7]. The main features are like immediate image access facility, doctorpatient direct interaction, good quality, resolution and speed of image transmission from remote region over wireless networks.
d
M
Abstract - The clinical diagnoses is increasingly getting dependant on chronic wound images for screening, diagnosis, treatment planning, and routine check-ups. This work aims proposing a methodology for development of a chronic wound image database which may facilitate treatment of wounds with the help of storeand-forward Telemedicine approach for accessing the status of chronic wounds. This may further help a patient to acquire timely advice of his/her medical problem. Our major objective is to development an integrated approach of free publicly available global chronic wound community database of chronic wound images for pressure, diabetic, arterial and venous ulcer and its annotation with the supporting metadata through Telemedicine platform. In this paper we describe a smart-phone integrated low cost and ever greater quality based metadata creation process for chronic wound image acquisition at patient side and also provide a smooth interaction between doctor and patients remotely, endto-end routine based diagnostic and maintaining patient history respectively. The paper describes a prototype system that demonstrates and tests the approach. The implementation and simulation is carried out using Hypertext preprocessor (PHP) and MySqL database.
I. INTRODUCTION
te
Index Terms - Tele-wound Monitoring, Wound Metadata, Wound Database
G
Ac ce p
ENERAL peoples have been facing major challenges to check-up the wounded portion to the best consultant or doctor remotely. It is estimated that more than 13 million peoples worldwide suffer from chronic wounds per year, the number of patients suffering is continuously raising due to ageing population. According to Indian epidemiological chronic wounds was reported as 4.5 per 1000 population whereas that of acute wounds was 10.5 per 1000 population [1]. Fig. 1 depicts the worldwide wound status in the wound care market in 2014. The author [3] investigates the automated pressure ulcer tool that can be integrated with Telemedicine system aiming to increase the efficiency and monitoring patient’s data but they are not clearly mentioned about the data handling and storage information. The user manually draws the border contour on the touch screen of the smart phones [4]. This paper [5] proposes a practical, fast, non-invasive and portable wound area measurement method for individual Manuscript submitted July 2, 2014. Chinmay Chakraborty is with the Birla Institute of Technology, Mesra, India, (e-mail: [email protected]). Bharat Gupta is with the Birla Institute of Technology, Mesra, India, (email: [email protected]). Soumya K. Ghosh is with the Indian Institute of Technology, Kharagpur, India, (e-mail: [email protected]).
Fig. 1: Worldwide wound incidence (in millions) [2]
The most commonly carried devices for people like advanced messaging devices, electronic diaries, palmtop and mobile phones are providing continuous support to improve healthcare. The author proposed real time Telemedicine based diabetes monitoring using application SMS, wired phone to glucometer and reminder diary technologies [8]. The author has been noted that there are no problems when patients used smart phones for clinical data exchange also reported low activity in transferring glucose values using smart phones [9]. The authors develop a novel mobile health (mHealth) model using high preference-smart phone apps to support complex self-care tasks, continuous adherence data to the clinician portal, to receive regimens and secure two-way communications between patients and clinicians [10]. Smart phones with cameras gives a promising metadata annotation [11] and sharing platform and also have inherent network connectivity, the user interfaces are developing at a fast rate. Good metadata should be authoritative, appropriate to patient’s clinical information collection, includes a clear statement and supports long term management respectively. The major role of designing powerful tool like metadata in networked smart-phones is that to acquire chronic wound images and patient’s demographic information in single page. Page 1 of 16
Mobile Metadata Assisted Community Database of Chronic Wound Images Chinmay Chakraborty, Bharat Gupta, and Soumya K. Ghosh Abstract - The clinical diagnoses is increasingly getting dependant on chronic wound images for screening, diagnosis, treatment
ip t
planning, and routine check-ups. This work aims proposing a methodology for development of a chronic wound image database which may facilitate treatment of wounds with the help of store-and-forward Telemedicine approach for accessing the status of chronic wounds. This may further help a patient to acquire timely advice of his/her medical problem. Our major objective is to development an
cr
integrated approach of free publicly available global chronic wound community database of chronic wound images for pressure, diabetic, arterial and venous ulcer and its annotation with the supporting metadata through Telemedicine platform. In this paper we
us
describe a smart-phone integrated low cost and ever greater quality based metadata creation process for chronic wound image acquisition at patient side and also provide a smooth interaction between doctor and patients remotely, end-to-end routine based diagnostic and maintaining patient history respectively. The paper describes a prototype system that demonstrates and tests the
an
approach. The implementation and simulation is carried out using Hypertext preprocessor (PHP) and MySqL database. Index Terms - Tele-wound Monitoring, Wound Metadata, Wound Database
M
I. INTRODUCTION
General peoples have been facing major challenges to check-up the wounded portion to the best consultant or doctor remotely. It is estimated that more than 13 million peoples worldwide suffer from chronic wounds per year, the number of patients suffering
d
is continuously raising due to ageing population. According to Indian epidemiological chronic wounds was reported as 4.5 per 1000 population whereas that of acute wounds was 10.5 per 1000 population [1]. Fig. 1 depicts the worldwide wound status in
te
the wound care market in 2014. The author [3] investigates the automated pressure ulcer tool that can be integrated with Telemedicine system aiming to increase the efficiency and monitoring patient’s data but they are not clearly mentioned about the
Ac ce p
data handling and storage information. The user manually draws the border contour on the touch screen of the smart phones [4]. This paper [5] proposes a practical, fast, non-invasive and portable wound area measurement method for individual patient but not focused on large number of patients data handling techniques. Portable devices like networked mobile camera phone can be used to capture the image because of it is a more powerful, low processing power device, handling complex data, presents graphical user interface (GUI) and user friendly in real-time manner at a distance [6] [7]. The main features are like immediate image access facility, doctor-patient direct interaction, good quality, resolution and speed of image transmission from remote region over wireless networks.
Fig. 1: Worldwide wound incidence (in millions) [2]
1
Page 2 of 16
The most commonly carried devices for people like advanced messaging devices, electronic diaries, palmtop and mobile phones are providing continuous support to improve healthcare. The author proposed real time Telemedicine based diabetes monitoring using application SMS, wired phone to glucometer and reminder diary technologies [8]. The author has been noted that there are no problems when patients used smart phones for clinical data exchange also reported low activity in transferring glucose values using smart phones [9]. The authors develop a novel mobile health (mHealth) model using high preference-smart
ip t
phone apps to support complex self-care tasks, continuous adherence data to the clinician portal, to receive regimens and secure two-way communications between patients and clinicians [10]. Smart phones with cameras gives a promising metadata annotation [11] and sharing platform and also have inherent network connectivity, the user interfaces are developing at a fast
cr
rate. Good metadata should be authoritative, appropriate to patient’s clinical information collection, includes a clear statement and supports long term management respectively. The major role of designing powerful tool like metadata in networked smart-
us
phones is that to acquire chronic wound images and patient’s demographic information in single page. This patient information centric mobile metadata can be transmitting to TMH remotely under Telemedicine framework. The main difference between data and metadata is that data deals with the patient’s clinical information only and metadata is able to gather the large volume of data
an
like wound images with clinical information. The high feature based APPLE, Nokia, SAMSUNG GALAXY TAB, Windows phone, BlackBerry, iOS, Canon, Adobe, ANDROID platforms with cloud service and GSM mobile phones can be used [12]. This author [13] has been proposed a GPS-driven location based semantic metadata for organizing and sharing images. The
M
automated mobile media metadata creation process has been developed [14].
The most of the doctors still utilize handwritten prescriptions. Current patient records are subject to many drawbacks, such as
d
missing and incomplete data entries by the end user, unavailability because of use and the general inability to be combined and
te
manipulated. Very few doctors are electronically linked up to Tele-medical hub (TMH) or pharmacies [15]. The total medication history has not been ever safely maintained. Data breakup occurs whenever a patient is seen by multiple doctors who
Ac ce p
independently prescribe without seeing previous history. The big challenge is that as the population ages the demand placed upon healthcare delivery will increase drastically as well as the number of attendant prescribed medications. Telemedicine deals networks of inpatient and outpatient facilities, including clinics, doctor offices and even home-healthcare settings, which are usually spread throughout a large geographical area. Unfortunately, there is no current industry standard data model for properly maintained medication history of the patient. The properly medication data model is required. Doctor then pays a visit to the database and finds a new entry for diagnosis. He picks up the patient identity number (PIDNUM), visits the patient’s profile and studies the report. After visiting the patient’s details then doctors able to send the e-prescribed feedback report. Patient can use the feedback and treat himself if possible according to the doctor’s feedback and if not he pays a visit to the nearest dermatologist. Often long-term care facilities do not have physicians on premise and health concerns can be beyond an onsite care giver’s scope. In these situations, traditional Telemedicine technology is not sufficient. Our proposed models helps to mitigate the health burden specially for rural people where best doctor’s are not available, Tele-medical agents (TMA) are responsible to collect the patient’s data (image and clinical data i.e. metadata format) with unique PIDNUM, sends to wounds database where large volume of data can be stored, then doctor’s can visit with respect to PIDNUM and feed to e-prescribed medication to TMAs or patient’s party. This model is able to reduce the face-to-face visits, transportation cost and provides good level of patient’s satisfaction.
2
Page 3 of 16
The remaining of this paper is organized as follows: section II explains the requirements of Tele-wound monitoring like programming platform, metadata format, patient centric information encoding, and doctor’s feedback and security mechanism respectively. Section III explains our proposed system model while section IV explains the clinical output results with screenshots and Section V provides a conclusion and future direction of this work. II. FRAMEWORK FOR TWM SYSTEM
ip t
Our proposed Tele-wound monitoring (TWM) system provides a telemedicine service intended at monitoring the wound status of patients remotely with certain interval and also reduce the frequency of clinic visits. The TWM is built on the concept that equips every person with ability to get diagnosed with little knowledge of internet and how smart-phones operate. The networked
cr
smart-phones having limited computational resources like lower processing speed and memory than computers. Smart-phone based Telemedicine overcomes the shortcomings of computer based Telemedicine like requirement not portable, wired service
us
and constant electricity. Tele-monitoring provides to monitor the patient’s vital signs regularly for reducing time and cost [16]. In this paper [17] compared the cost of electronic data transmission to the cost of routine clinic visits. Clinical information can be collected manually by TMA with high resolution based cell-phone that may be used to optimize the patient's monitoring and
an
treatment protocols. The good quality wounded image has been chosen by the cell-phones and is sent to the TMH and stored in wound database with unique PIDNUM for monitoring the wound status problem remotely. Fig. 2 depicts proposed TWM system
Ac ce p
te
d
M
configuration.
Fig. 2: Proposed Tele-wound Monitoring (TWM) System Configuration
A. Platform
The software which is installed in smart-phone processes the metadata and forwards them via Internet to wound database server in TMH. The several language based platforms is developed smart-phone software. TWM is a major challenging part which has got promising future. Hence the application made for it and the database designed has vivid base. Risto et al. [18] proposed Symbian C++ application platform instead of Java 2 mobile edition (J2ME) because J2ME does not provide sufficient access to low level resources. The following databases are used for metadata storage along with images like ORACLE, SQL, MS ACCESS, XML (Extended mark-up language), SAS8, PHP/MYADMIN, but most of it very costly to implement and licensed. Paul A. et al. proposed metadata driven software application for research electronic data capture using PHP+, JavaScript languages and a MySQL database engine for data storage and manipulation [19]. The proposed global wound database is handled by MySQL in ISM engine. PHP/MySQL is open source and very popular choice of database. PHP and MySQL are very compatible when used together and give various dynamic functionalities and are run easily on almost all the 3
Page 4 of 16
platforms. Application for TWM is made on PHP, HTML, CSS and JAVASCRIPT platform. The simple scripting language like PHP is used to designed for generation of dynamic web pages and easy to embedded into HTML source document. HTML web based smart-phone interface enables doctors and patients to interact, access and capture wound images at 24/7 from diverse location, while clinically indexing comprehensive metadata to the TMH. Database entries and queries are brought about very easily. To start with highly skilled and professional doctors are approached and they create their account in TWM applications or apps for accessing the database. The implemented patient profile on mobile metadata architecture has been shown below in Fig.
an
us
cr
ip t
3.
M
Fig. 3: Mobile-metadata Architecture
B. Metadata Format
Two different approaches used for the representation of images are: (a) metadata based and (b) content based. The metadata
d
(both automated and user-assisted) would define a transformation to convert a source variable to a target variable and or data
te
about data and also is designed to support the discovery, access, usability, documents, management and preservation of data to improve wound healing. The transforms metadata would contain the mapping of variables from one data standard to another data
Ac ce p
standard. There are no reports available on community wound database development. Picture archiving and communication systems (PACS) are responsible for managing the image data but not managing or querying image metadata during interpretation [12]. The electronic storage unit stores the patient’s personal data, demographic information, clinical information and digital wounded images. The medical information contains patient age, gender, wound site, previous visit history, wound images, computer generated reports, clinician’s diagnosis etc. The data can be classified as: (a) Patient’s personal data (i.e. PIDNUM, Name, Address, Date of birth, Birth place, Sex etc), (b) Patient’s medical data (i.e. Plain text, Image, Textual, Video etc), (c) Medical expert’s personal data (i.e. doctor’s personal information, Unique identification code) and (d) System management related data (i.e. Patient’s list, Password files, log files etc). The metadata based patient profile with standalone and PIDNUM based format has been shown in Fig. 4.
4
Page 5 of 16
Metadata Wound Image
Patient’s Clinical Information
Upload Name Image 1
ID Sex
Wound Site Previous Visit History Parents History Current Medications
us
Ethnic Origin
cr
Address at time of Diagnosis
ip t
D.O.B Image 2
Telephone Number Year of Diagnosis
an
Comments
Timer: (data/time: acquired and saved successfully) Fig. 4: Metadata Format for Patient Profile Creation
M
C. Patient Profile/ Client end Approach
A chronic wound effected patient needs and wants a remote care, make contact with TMAs who are having networked camera based smart-phones; it creates his account in the apps and feeds the form with relevant clinical information. Every patient has his
d
profile once accounts create. Patients are asked to feed their name, wound site, wound images and also some comment relating to
te
their wound after which they are redirected to their profile and a PIDNUM is generated in the wound database which is unique for every patient. The PIDNUM has to keep in mind to the patient’s for future treatment, no need to carry hardcopy multiple
Ac ce p
prescriptions. Every detail of the patient is saved in the global wound database. Profile of patients shows their personal information including wound images and also link to profiles of doctors who have their account in TWM centre. This database stored registered patient’s detail and authorized best dermatologists information, so agents can help to show the details information to select the doctors. Before a patient selects a doctor for diagnosis of his wound he visits doctor’s profile and checks out information such as his qualification, experience and number of cases solved. It gives a clear picture to the patient whom to select for their diagnosis. Our proposed model introduces automatic query system based on wound type which automatically suggests the best doctor’s in their disease. There is a button on the bottom part of every doctor’s profile clicking on which the report of the patient is sent to the concerned doctor. The PIDNUM is saved in the database of that particular doctor. It takes some time for the patient to get e-prescribed medication and feedback from the doctor. Fig. 5 depicts the screenshot of patient centric data collection by TMA using Android operating system based Xolo Q1000 Opus smart-phone.
5
Page 6 of 16
cr
ip t Ac ce p
te
d
M
an
us
(a) Patient’s Clinical Information
(b) Patient’s wound Image
Fig. 5: Patient’s Data Collection by TMA through Mobile Interface
D. Doctor’s Profile/ Server end Approach
Expert doctors keep on checking their profile for reports from patients. When a doctor logs in the apps and is redirected to his profile he finds that he has a case which asks for care and diagnosis. He clicks on the link on the arrived case which says “visit”. This redirects doctor to a page ethic shows all the fed details of patients. Analysis of pictures of wound site and comments of patient makes clear the intensity of injury and seriousness of wound. Now the doctor has two options. Either he can go ahead and analyse the wound or can refer the report to other existing doctors on the apps. In case the report is referred, PIDNUM of that particular patient goes in the database of the doctor to whom the report is forwarded. If doctor chooses to diagnose the patient himself he goes in to analyse the report even more deeply by segmenting the wound site picture and sends the picture for colour correction in computer added diagnostic (CAD) tool. The image processing is also being there at server end in CAD tool. Once the wound site images are colour corrected and segmented based on the colour, texture and tissue type study of wound site reveals the type of wound patient has. Doctor now goes ahead and feeds in some prescription and feedback. Once the patient is treated and doctor makes it sure that the patient is treated he clicks on the “treated” button and it increases the cases solved by 6
Page 7 of 16
that particular doctor counter increment by 1. The report of the patient goes in the solved cases section of the doctor. At later stage whenever doctor finds it necessary to check out the cases solved by him he goes ahead and from the drop down menu on his profile which has two links “present cases” and solved cases. Solved cases link gives the report of the patient who are already treated by the doctor. Whenever a patient is treated the flag in doctor’s database is reset and it goes in solved cases list. Patients keep on checking their profile after they send their report to doctors. Doctors regularly keep sending prescriptions and feedbacks which are showed on the “doctor’s feedback” screen on patient’s profile. It helps patients who are in remote areas and do not
ip t
have access to expert care and advice. There is also a provision that patients can update their report by clicking on “update” button present on their profile. By updating their report patients can keep doctors updated with their status of wound and it can be brought about by either dropping in comments or uploading pictures of wound site. An update from patient is reflected in the
cr
database and a flag in doctor’s database is also set which reveals that report of the particular patient needs to be reviewed. This TWM apps helps people in remote sites of the world to have expert advice from doctors who are highly skilled and professional.
an extra burden. Fig. 6 shows the prototype implementation from doctor side.
an
E. Wound Database
us
It also helps elderly people to have home care in case of periodic checkups for whom transportation from remote site to clinic is
The database would be even stronger to provide the patient and doctor with even more dynamic features that allows more doctor-patient smooth interaction and vice-versa. If after diagnosis the doctor finds that site of wound has reached threshold
M
level, then will ask the patient to visit him or any other doctor nearby to the patient’s location which is judged from the address of the patient. The patient is also registered in the database of the visiting patient and given a date and time slot subject to availability. This is a favor which doctor will do on the patient. The date and time slot will be texted to the patient on his
d
telephone number and will also be updated on his profile. Medetec wound image database [20] has been used for this work. The medical image databases like Cumulative Index to Nursing and Allied Health Literature (CINAHL), Medline, nursing and
te
medical online journals, the British Nursing Index (BNI), The Cochrane Library, internet sources, and Department of Health
Ac ce p
(DoH) databases were searched for accessing the wound images.
(a)
7
Page 8 of 16
cr
ip t te
d
M
an
us
(b)
(c)
Ac ce p
Fig. 6 (a) (b) (c): Prototype Implementation with Patient’s ID and Feedback Panel
F. Metadata Protection
The major security issues will arise when transferring the metadata from smart-phone to TMH. Privacy and security both of are the important in any healthcare information systems. Patient’s clinical information should be always being considered as sensitive in general and identifying data should remain strictly confidential. Security mechanisms are required to protect the patient data, prevent improper system use and to insure the confidentiality. Each patient is assigned a unique PIDNUM and password for system sign-on and use. The mHealth [21] mainly use of mobile and wireless technology that enables doctors to remotely monitor patients health regularly, managing patient care, reducing drug shortages at health clinics, improve quality health care unit that’s why need to generate novel security and privacy issues [22], reducing cost and allowing patients to spend less time in the clinic. According to National Committee for Vital and Health Statistics (NCVHS), PHI as ‘personal health information’ rather than ‘protected health information’ which is an axiom that has specific meaning in a HIPAA perspective [23]. III. PROTOTYPE WOUND MONITORING SYSTEM This developed platform allows doctors to measure the progress of a wound timely. The paper makes an attempt to make a system that can efficient tool using PHP/MySqL simulation. Patient’s clinical information and wound images are linking with metadata and embedded in JPEG that is globally accessible through a web application. Metadata is also stored in the MySQL 8
Page 9 of 16
database. The number of times the patient visits his profile and updates something it is maintained in the database. And after every update the seen flag will be reset to “Not Visited” and that prompts the doctor to revisit the patient’s profile and rediagnose the report and send another feedback. The key use cases include: (a) Update patient clinical status, (b) Store wound images time-to-time to the database, (c) Update privacy code, (d) Delete arbitrary registry metadata and (e) Unchanged the PIDNUM. The Data required for this paper work was chronic wound patient information collected from a Medetec database. The chronic wound patient dataset contains around 100 records from a clinical computer system. The records are patient’s summaries
ip t
of various stages of wound images in the form of HTML pages. The portable devices like smart-phones, tablets, laptops with inbuilt camera capture wound images and computing and networking features allow smooth interaction. The MySqL/PHP based
cr
specific implementation staffs are presented below:
Step 1: Creation of patient’s demographic page using following commands
us
Step 2:
$fpass=$_POST['ide']; //PIDNUM is posted from form to the login page and is saved in
mysql_connect('localhost','root','')or die(mysql_error()); //mysql server is started mysql_select_db('medical')or die(mysql_error());
d
// concerned database is selected
M
$fpass variable
an
//beginning of php code
te
$result=mysql_query("SELECT PIDNUM FROM account")or die(mysql_error()); //id of all the patients are selected and saved in an array
Ac ce p
while($row=mysql_fetch_array($result)){ if($row['id']==$fpass){
// if PIDNUM of the current user matches with that present in array then the user is redirected to his own very profile else he/she is asked to create account $bunt=1;
$myfile="login.txt";
// retrieves the PIDNUM of the current patient $f=fopen($myfile,'w')or die("can't open file"); fwrite($f,$fpass); fclose($f); break; } else{ $bunt=0; } } ?>
9
Page 10 of 16
//end of php code Step 3: if ($success) redirect to patient’s profile else echo “create your account”
ip t
redirect to sign up page
cr
A. Flowchart:
The encoded data format should be sent from smart phone to TMH using Internet. The general person can get the doctors
us
details from TMA but not having permission to access the database or create a bulk doctor profile. The first step is SignUp as patient. Any person who wants to get expert advice from a doctor about his wound or wants his wound to get diagnosed creates
an
his account wherein he has to feed in some details along with details of wounded images area. The completion of these steps redirects him to his profile and from his profile page he can select a doctor for the diagnosis purpose. Another link on the page says sign in as patient. Someone who already has account on this application has unique PIDNUM and using that he tries to
M
login. If the login PIDNUM is valid he is redirected to profile page else it says not a registered user and he is asked to create account. On the profile page patient finds feedback and prescribed report from doctor in case his report is visited by doctors and seen. The third link on the application home page says sign in as doctor. This link is clicked by a doctor to visit his profile
d
wherein he finds reports from patients from various places and waiting to be visited and diagnosed. If the doctor finds case to be well and good enough to be handled by him he visits and goes ahead for diagnosis else he refers that particular case to other
Ac ce p
te
doctors catering to TWM application. Fig. 7 depicts the flow-map of TWM system.
Fig. 7: Tele-wound Monitoring Flow-map
10
Page 11 of 16
IV. RESULTS Every patient is asked to feed the following fields: clinical information like name, age, sex, date of birth, address and contact details etc and also taken wound images using smart-phones by TMA and send to TMH for best daigonosis. The submitted patient report stored in a global wound database with specified PIDNUM that serves as the key for retrieving the details from database. Fig. 8 has shown that the screenshot of doctor accessing the patient detail’s.
ip t
Patients Profile:
A. Login/Sign up – Patient’s information protection is big concern for maintaning confidentiality to enter a username and password.
cr
B.Chronic wound list – The location of wound effected portion is used to organize the wound list. The TMA easily see the
te
d
M
an
us
different types of wounds.
Fig. 8: Screenshot of wound lists in smart phone
PIDNUM.
Ac ce p
C. Patient history – The wound image and clinical information of the patient has been stored in the database with unique
D. Treatment – It allows a user to input information regarding any changes in wound images after getting prescribed medication by doctor. The doctor can visit patient’s images over a certain time interval. The previous treatment can be accesssed from patient history.
E. Wound History – The global community database has been developed to store patient’s information and as well as maintain the doctor’s profile.
The Fig. 9 depicts the screenshot of doctor’s accessing the patient details. The Fig. 10 below shows the screenshot of doctor’s seeing medical database including patient’s clinical information and wound images.
11
Page 12 of 16
ip t cr us
Ac ce p
te
d
M
an
Fig. 9: Screenshot of Doctor’s accessing the Patient Detail’s
Fig. 10: Screenshot of Doctor’s Profile
V. CONCLUSION AND FUTURE SCOPE This paper presents a web-enabled flexible informatics systems based framework for analyzing and processing wound images. The wound images have been taken from patient body through high resolution camera assisted networked mobile phones and stored in metadata (having efficient saving scheme) format and send it to CAD tool in TMH by the TMA remotely. The patients can get prescribed medication from the doctors with respect to unique PIDNUM. This approach gives a better response in terms of unique PIDNUM creation; patient can move one doctor to another having this unique PIDNUM, so full patient’s history easily 12
Page 13 of 16
accessible by several doctors. The CAD tool having image processing toolbox but it is not setup at client sides because of multiple number of agents are sending the patient’s details that embed image and clinical information to the global community database and cost is also highly demanding. It gives a clinically effectiveness, highly efficient, timely, more comfort and stress, reduced need for hospitalization and user friendly performances. The global wound database would be even stronger to provide the patient and doctor with even more dynamic features that allows more doctor-patient interaction vice-versa. In future we will propose an algorithm for Tele-truma case where patients can upload their images and videos of wound for better treatment. The
ip t
advanced metadata protection mechanisms are required for sharing wound images. The big challenges in managing large volumes of medical data that can be integrate with electronic health records for big data analytics purpose. This novel approach
cr
can be used in data mining application and public health monitoring systems. REFERENCES
us
[1]Gupta N, Gupta SK, Shukla VK and Singh SP (2004), “An Indian community based epidemiological study of wounds”, Journal of Wound Care, vol. 13, no. 8, pp. 323-325, Sept. 2004.
[2]Big changes coming to the wound care market. (2014). Available: http://seekingalpha.com/article/1757332-big-changes-coming-to-the-wound-care-marketin-2014.
an
[3]Dimitrios I. K and Fotini L. T, “Automated pressure ulcer lesion diagnosis for Telemedicine systems”, IEEE Engineering in Medicine and Biology Magazine, vol. 26, issue 5, pp. 18-22, Oct. 2007.
[4]Kass M, Witkin A,Terzopulos D.,”Snakes: Active contour models”, International Journal of Computer Vision, pp. 321-331, 1988.
M
[5]Hettiarachchi N. D. J, Mahindaratne R. B. H, Mendis G. D.C. and Nanayakkara H. T.,” Mobile based wound measurement”, IEEE Point-of-Care Healthcare Technologies (PHT), Bangalore, India, pp.298-301, Jan. 16-18, 2013.
[6]Orlando R.E. Pereira, Paulo A.C.S. Neves, Joel J.P.C. Rodrigues, “Mobile solution for three-tier biofeedback data acquisition and processing”, Proc. of IEEE
d
GlOBECOM, Global Telecom. Conf., New Orleans, LO, pp.1-5, Nov. 2008.
te
[7]Mawston N., “Camera phones outsell digital still cameras in H1 2003 and beyond”, Wireless devices strategies, Strategy Analytics, Newton Centre, MA, 2003.
Ac ce p
[8]Farmer A, Gibson O, Hayton P, et al., “A real-time, mobile phone-based telemedicine system to support young adults with type 1 diabetes”, Inform Primary Care, vol. 13, pp. 171–177, 2005.
[9]Vahatalo M.A., Virtamo H.E., Viikari J.S., Ronnemaa T., “Cellular phone transferred self blood glucose monitoring: prerequisites for positive outcome”, Practical Diabetes International, vol. 21, no. 5, pp. 192 – 194, 2004. [10] Bambang Parmanto, Gede Pramana, Daihua Xie Yu, Andrea D Fairman, Brad E Dicianno, Michael P McCue, “iMHere: A Novel mHealth system for supporting self-care in management of complex and chronic conditions”, Journal of Medical Internet Research Mhealth Uhealth, vol. 1, iss. 2, pp. 1-16, Nov., 2013.
[11] Gargi U., “Modelling and clustering of photo capture streams”, Proc. of Multimedia, ACM Press, pp.47-54, 2003. [12] Malcolm S, “Smartphone apps, Permissions and privacy concerns and next steps”, TEKDESK: A division of the community opportunity and innovation network, pp. 1-94, July, 2013. [13] Toyama, K., Logan, R., Roseway, A., Anandan, P.Geographic Location tags on digital images, Proceedings of the 11th ACM International Conference on Multimedia, ACM Press; pp.156-166. [14] Risto S., Eric H., Anita W. and Marc D., “Metadata creation system for mobile images”, ACM MobiSys’04 Proc. of the 2nd Int. Conf. on Mobile systems, applications, and services, Boston, USA, pp. 36-48, June, 2004.
13
Page 14 of 16
[15] Chinmay C., Bharat G. and Soumya K. G., “A review on telemedicine-based WBAN framework for patient monitoring”, Telemedicine and e-health, Marry Ann Liebert, Inc, vol. 19, no. 8, pp. 619-626, Aug., 2013. [16] Stephane Meystre, “The current state of Telemonitoring: A comment on the literature”, Telemedicine and e-Health, Mary Ann Liebert Inc., vol. 11, no. 1, pp. 63-69, Feb., 2005. [17] Chase, H. P., Pearson, J. A., Wightman, C., Roberts, M. D., Oderberg, A. D. & Garg, S. K., “Modem transmission of glucose values reduces the cost and need for clinic visits”, Diabetes Care, vol. 26, no. 5, pp. 1475–1479, May, 2003.
ip t
[18] Risto S., Mikko V., Juha P. and Hanno N., “Mobshare: Controlled and immediate sharing of mobile images”, MULTIMEDIA '04 Proc. of the 12th Annual ACM Int. Conf. on Multimedia, USA, pp.724-731, Oct., 2004.
cr
[19] Paul A. Harris, Robert Taylor, Robert Thielke, Jonathon Payne, Nathaniel Gonzalez and Jose G. Conde, “Research electronic data capture (REDCap) - A metadata-driven methodology and workflow process for providing translational research informatics support”, Journal of Biomedical Inform., Vol. 42, no.
us
2, pp. 377–381. doi:10.1016/j.jbi.2008.08.010, April, 2009. [20] Medetec wound database: Available: http://www.medetec.co.uk/files/medetec_image_databases.html.
[21] World health Organization, “mHealth: New horizons for health through mobile technologies: second global survey on eHealth”, Global Observatory for
an
eHealth Series, Vol. 3, pp. 1-112, 2011.
[22] Al Ameen M, Liu J, Kwak K, “Security and privacy issues in wireless sensor networks for healthcare applications”, Journal of Medical System, Springer,
M
vol. 36, no. 1, pp. 93-101, March, 2010.
[23] Choi Y.B., Capitan K.E., Krause J.S and Streeper M.M, “Challenges associated with privacy in healthcare industry, implementation of HIPAA and security
d
rules”, Journal of Medical System, vol. 30, no. 1, pp. 57-64, Feb. 2006.
te
Chinmay Chakraborty (M’09) was born in Sabang, West Midnapur, India in 1984. He received the B.Tech. degree in Electronics and Communication Engineering from West Bengal University of Technology
Ac ce p
(WBUT), India in 2006. From 2006 to 2007, he was with the Dept. of Industrial Engineering and Management, Indian Institute of Technology, Kharagpur (IIT Kharagpur), India as a Research Consultant. He received the M.S. degree in Telecommunication Engineering from G. S. Sanyal School of Telecommunications, IIT Kharagpur, India in 2010. He is worked at ICFAI Univ., Agartala, India as Senior Lecturer. Currently He is working at Birla Institute of Technology, Mesra, India as Assistant Professor. His research interests include Wireless Body Sensor Networks, Telemedicine, and Communication Networks.
Bharat Gupta received B.E and M.Tech degrees in Electronics Engineering from Govt. Engineering College of Ujjain, Madhya Pradesh, India and Birla Institute of Technology in 2000 and 2003 respectively. The Ph.D. degree he received in the area of Wireless Communication in Health care Monitoring from the Department of Microelectronics and Telecommunication at Univ. of Rome, Tor Vergata, Italy, in 2011. He has been teaching at the Birla Institute of Technology, Mesra, since August 2004 where he is presently an Associate Professor of the Electronics and Communication Engineering. His teaching and research interests include FM UWB Communication, Energy Efficient MAC for WBAN, Vital Sign Monitoring and WSN. He has published more than 20 technical papers in various journals and. He is a member of GTTI.
14
Page 15 of 16
Soumya K. Ghosh is Professor in the School of Information Technology, Indian Institute of Technology, Kharagpur (IIT Kharagpur), India. His primary areas of research include Geospatial Databases and Services, Cloud Computing and Security. Prior to IIT Kharagpur, he worked for Indian Space Research Organization in
ip t
area of remote sensing and geographic Information systems for natural resource management. He did his PhD and M.Tech in Computer Science, from Department of Computer Science & Engineering, IIT Kharagpur. He did his BE in Electronics and Communication Engineering from National Institute of Technology (formerly, Regional Engineering College),
Ac ce p
te
d
M
an
us
cr
Durgapur, India. He has published more than 100 technical papers in various journals and conferences.
15
Page 16 of 16
S2213-9095(14)00044-5 http://dx.doi.org/doi:10.1016/j.wndm.2014.09.002 WNDM 45
To appear in: Received date: Accepted date:
11-7-2014 29-9-2014
Please cite this article as: Chakraborty C, Gupta B, Ghosh SK, Mobile Metadata Assisted Community Database of Chronic Wound Images, Wound Med. (2014), http://dx.doi.org/10.1016/j.wndm.2014.09.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Mobile Metadata Assisted Community Database of Chronic Wound Images Chinmay Chakraborty, Bharat Gupta, and Soumya K. Ghosh
an
us
cr
ip t
patient but not focused on large number of patients data handling techniques. Portable devices like networked mobile camera phone can be used to capture the image because of it is a more powerful, low processing power device, handling complex data, presents graphical user interface (GUI) and user friendly in real-time manner at a distance [6] [7]. The main features are like immediate image access facility, doctorpatient direct interaction, good quality, resolution and speed of image transmission from remote region over wireless networks.
d
M
Abstract - The clinical diagnoses is increasingly getting dependant on chronic wound images for screening, diagnosis, treatment planning, and routine check-ups. This work aims proposing a methodology for development of a chronic wound image database which may facilitate treatment of wounds with the help of storeand-forward Telemedicine approach for accessing the status of chronic wounds. This may further help a patient to acquire timely advice of his/her medical problem. Our major objective is to development an integrated approach of free publicly available global chronic wound community database of chronic wound images for pressure, diabetic, arterial and venous ulcer and its annotation with the supporting metadata through Telemedicine platform. In this paper we describe a smart-phone integrated low cost and ever greater quality based metadata creation process for chronic wound image acquisition at patient side and also provide a smooth interaction between doctor and patients remotely, endto-end routine based diagnostic and maintaining patient history respectively. The paper describes a prototype system that demonstrates and tests the approach. The implementation and simulation is carried out using Hypertext preprocessor (PHP) and MySqL database.
I. INTRODUCTION
te
Index Terms - Tele-wound Monitoring, Wound Metadata, Wound Database
G
Ac ce p
ENERAL peoples have been facing major challenges to check-up the wounded portion to the best consultant or doctor remotely. It is estimated that more than 13 million peoples worldwide suffer from chronic wounds per year, the number of patients suffering is continuously raising due to ageing population. According to Indian epidemiological chronic wounds was reported as 4.5 per 1000 population whereas that of acute wounds was 10.5 per 1000 population [1]. Fig. 1 depicts the worldwide wound status in the wound care market in 2014. The author [3] investigates the automated pressure ulcer tool that can be integrated with Telemedicine system aiming to increase the efficiency and monitoring patient’s data but they are not clearly mentioned about the data handling and storage information. The user manually draws the border contour on the touch screen of the smart phones [4]. This paper [5] proposes a practical, fast, non-invasive and portable wound area measurement method for individual Manuscript submitted July 2, 2014. Chinmay Chakraborty is with the Birla Institute of Technology, Mesra, India, (e-mail: [email protected]). Bharat Gupta is with the Birla Institute of Technology, Mesra, India, (email: [email protected]). Soumya K. Ghosh is with the Indian Institute of Technology, Kharagpur, India, (e-mail: [email protected]).
Fig. 1: Worldwide wound incidence (in millions) [2]
The most commonly carried devices for people like advanced messaging devices, electronic diaries, palmtop and mobile phones are providing continuous support to improve healthcare. The author proposed real time Telemedicine based diabetes monitoring using application SMS, wired phone to glucometer and reminder diary technologies [8]. The author has been noted that there are no problems when patients used smart phones for clinical data exchange also reported low activity in transferring glucose values using smart phones [9]. The authors develop a novel mobile health (mHealth) model using high preference-smart phone apps to support complex self-care tasks, continuous adherence data to the clinician portal, to receive regimens and secure two-way communications between patients and clinicians [10]. Smart phones with cameras gives a promising metadata annotation [11] and sharing platform and also have inherent network connectivity, the user interfaces are developing at a fast rate. Good metadata should be authoritative, appropriate to patient’s clinical information collection, includes a clear statement and supports long term management respectively. The major role of designing powerful tool like metadata in networked smart-phones is that to acquire chronic wound images and patient’s demographic information in single page. Page 1 of 16
Mobile Metadata Assisted Community Database of Chronic Wound Images Chinmay Chakraborty, Bharat Gupta, and Soumya K. Ghosh Abstract - The clinical diagnoses is increasingly getting dependant on chronic wound images for screening, diagnosis, treatment
ip t
planning, and routine check-ups. This work aims proposing a methodology for development of a chronic wound image database which may facilitate treatment of wounds with the help of store-and-forward Telemedicine approach for accessing the status of chronic wounds. This may further help a patient to acquire timely advice of his/her medical problem. Our major objective is to development an
cr
integrated approach of free publicly available global chronic wound community database of chronic wound images for pressure, diabetic, arterial and venous ulcer and its annotation with the supporting metadata through Telemedicine platform. In this paper we
us
describe a smart-phone integrated low cost and ever greater quality based metadata creation process for chronic wound image acquisition at patient side and also provide a smooth interaction between doctor and patients remotely, end-to-end routine based diagnostic and maintaining patient history respectively. The paper describes a prototype system that demonstrates and tests the
an
approach. The implementation and simulation is carried out using Hypertext preprocessor (PHP) and MySqL database. Index Terms - Tele-wound Monitoring, Wound Metadata, Wound Database
M
I. INTRODUCTION
General peoples have been facing major challenges to check-up the wounded portion to the best consultant or doctor remotely. It is estimated that more than 13 million peoples worldwide suffer from chronic wounds per year, the number of patients suffering
d
is continuously raising due to ageing population. According to Indian epidemiological chronic wounds was reported as 4.5 per 1000 population whereas that of acute wounds was 10.5 per 1000 population [1]. Fig. 1 depicts the worldwide wound status in
te
the wound care market in 2014. The author [3] investigates the automated pressure ulcer tool that can be integrated with Telemedicine system aiming to increase the efficiency and monitoring patient’s data but they are not clearly mentioned about the
Ac ce p
data handling and storage information. The user manually draws the border contour on the touch screen of the smart phones [4]. This paper [5] proposes a practical, fast, non-invasive and portable wound area measurement method for individual patient but not focused on large number of patients data handling techniques. Portable devices like networked mobile camera phone can be used to capture the image because of it is a more powerful, low processing power device, handling complex data, presents graphical user interface (GUI) and user friendly in real-time manner at a distance [6] [7]. The main features are like immediate image access facility, doctor-patient direct interaction, good quality, resolution and speed of image transmission from remote region over wireless networks.
Fig. 1: Worldwide wound incidence (in millions) [2]
1
Page 2 of 16
The most commonly carried devices for people like advanced messaging devices, electronic diaries, palmtop and mobile phones are providing continuous support to improve healthcare. The author proposed real time Telemedicine based diabetes monitoring using application SMS, wired phone to glucometer and reminder diary technologies [8]. The author has been noted that there are no problems when patients used smart phones for clinical data exchange also reported low activity in transferring glucose values using smart phones [9]. The authors develop a novel mobile health (mHealth) model using high preference-smart
ip t
phone apps to support complex self-care tasks, continuous adherence data to the clinician portal, to receive regimens and secure two-way communications between patients and clinicians [10]. Smart phones with cameras gives a promising metadata annotation [11] and sharing platform and also have inherent network connectivity, the user interfaces are developing at a fast
cr
rate. Good metadata should be authoritative, appropriate to patient’s clinical information collection, includes a clear statement and supports long term management respectively. The major role of designing powerful tool like metadata in networked smart-
us
phones is that to acquire chronic wound images and patient’s demographic information in single page. This patient information centric mobile metadata can be transmitting to TMH remotely under Telemedicine framework. The main difference between data and metadata is that data deals with the patient’s clinical information only and metadata is able to gather the large volume of data
an
like wound images with clinical information. The high feature based APPLE, Nokia, SAMSUNG GALAXY TAB, Windows phone, BlackBerry, iOS, Canon, Adobe, ANDROID platforms with cloud service and GSM mobile phones can be used [12]. This author [13] has been proposed a GPS-driven location based semantic metadata for organizing and sharing images. The
M
automated mobile media metadata creation process has been developed [14].
The most of the doctors still utilize handwritten prescriptions. Current patient records are subject to many drawbacks, such as
d
missing and incomplete data entries by the end user, unavailability because of use and the general inability to be combined and
te
manipulated. Very few doctors are electronically linked up to Tele-medical hub (TMH) or pharmacies [15]. The total medication history has not been ever safely maintained. Data breakup occurs whenever a patient is seen by multiple doctors who
Ac ce p
independently prescribe without seeing previous history. The big challenge is that as the population ages the demand placed upon healthcare delivery will increase drastically as well as the number of attendant prescribed medications. Telemedicine deals networks of inpatient and outpatient facilities, including clinics, doctor offices and even home-healthcare settings, which are usually spread throughout a large geographical area. Unfortunately, there is no current industry standard data model for properly maintained medication history of the patient. The properly medication data model is required. Doctor then pays a visit to the database and finds a new entry for diagnosis. He picks up the patient identity number (PIDNUM), visits the patient’s profile and studies the report. After visiting the patient’s details then doctors able to send the e-prescribed feedback report. Patient can use the feedback and treat himself if possible according to the doctor’s feedback and if not he pays a visit to the nearest dermatologist. Often long-term care facilities do not have physicians on premise and health concerns can be beyond an onsite care giver’s scope. In these situations, traditional Telemedicine technology is not sufficient. Our proposed models helps to mitigate the health burden specially for rural people where best doctor’s are not available, Tele-medical agents (TMA) are responsible to collect the patient’s data (image and clinical data i.e. metadata format) with unique PIDNUM, sends to wounds database where large volume of data can be stored, then doctor’s can visit with respect to PIDNUM and feed to e-prescribed medication to TMAs or patient’s party. This model is able to reduce the face-to-face visits, transportation cost and provides good level of patient’s satisfaction.
2
Page 3 of 16
The remaining of this paper is organized as follows: section II explains the requirements of Tele-wound monitoring like programming platform, metadata format, patient centric information encoding, and doctor’s feedback and security mechanism respectively. Section III explains our proposed system model while section IV explains the clinical output results with screenshots and Section V provides a conclusion and future direction of this work. II. FRAMEWORK FOR TWM SYSTEM
ip t
Our proposed Tele-wound monitoring (TWM) system provides a telemedicine service intended at monitoring the wound status of patients remotely with certain interval and also reduce the frequency of clinic visits. The TWM is built on the concept that equips every person with ability to get diagnosed with little knowledge of internet and how smart-phones operate. The networked
cr
smart-phones having limited computational resources like lower processing speed and memory than computers. Smart-phone based Telemedicine overcomes the shortcomings of computer based Telemedicine like requirement not portable, wired service
us
and constant electricity. Tele-monitoring provides to monitor the patient’s vital signs regularly for reducing time and cost [16]. In this paper [17] compared the cost of electronic data transmission to the cost of routine clinic visits. Clinical information can be collected manually by TMA with high resolution based cell-phone that may be used to optimize the patient's monitoring and
an
treatment protocols. The good quality wounded image has been chosen by the cell-phones and is sent to the TMH and stored in wound database with unique PIDNUM for monitoring the wound status problem remotely. Fig. 2 depicts proposed TWM system
Ac ce p
te
d
M
configuration.
Fig. 2: Proposed Tele-wound Monitoring (TWM) System Configuration
A. Platform
The software which is installed in smart-phone processes the metadata and forwards them via Internet to wound database server in TMH. The several language based platforms is developed smart-phone software. TWM is a major challenging part which has got promising future. Hence the application made for it and the database designed has vivid base. Risto et al. [18] proposed Symbian C++ application platform instead of Java 2 mobile edition (J2ME) because J2ME does not provide sufficient access to low level resources. The following databases are used for metadata storage along with images like ORACLE, SQL, MS ACCESS, XML (Extended mark-up language), SAS8, PHP/MYADMIN, but most of it very costly to implement and licensed. Paul A. et al. proposed metadata driven software application for research electronic data capture using PHP+, JavaScript languages and a MySQL database engine for data storage and manipulation [19]. The proposed global wound database is handled by MySQL in ISM engine. PHP/MySQL is open source and very popular choice of database. PHP and MySQL are very compatible when used together and give various dynamic functionalities and are run easily on almost all the 3
Page 4 of 16
platforms. Application for TWM is made on PHP, HTML, CSS and JAVASCRIPT platform. The simple scripting language like PHP is used to designed for generation of dynamic web pages and easy to embedded into HTML source document. HTML web based smart-phone interface enables doctors and patients to interact, access and capture wound images at 24/7 from diverse location, while clinically indexing comprehensive metadata to the TMH. Database entries and queries are brought about very easily. To start with highly skilled and professional doctors are approached and they create their account in TWM applications or apps for accessing the database. The implemented patient profile on mobile metadata architecture has been shown below in Fig.
an
us
cr
ip t
3.
M
Fig. 3: Mobile-metadata Architecture
B. Metadata Format
Two different approaches used for the representation of images are: (a) metadata based and (b) content based. The metadata
d
(both automated and user-assisted) would define a transformation to convert a source variable to a target variable and or data
te
about data and also is designed to support the discovery, access, usability, documents, management and preservation of data to improve wound healing. The transforms metadata would contain the mapping of variables from one data standard to another data
Ac ce p
standard. There are no reports available on community wound database development. Picture archiving and communication systems (PACS) are responsible for managing the image data but not managing or querying image metadata during interpretation [12]. The electronic storage unit stores the patient’s personal data, demographic information, clinical information and digital wounded images. The medical information contains patient age, gender, wound site, previous visit history, wound images, computer generated reports, clinician’s diagnosis etc. The data can be classified as: (a) Patient’s personal data (i.e. PIDNUM, Name, Address, Date of birth, Birth place, Sex etc), (b) Patient’s medical data (i.e. Plain text, Image, Textual, Video etc), (c) Medical expert’s personal data (i.e. doctor’s personal information, Unique identification code) and (d) System management related data (i.e. Patient’s list, Password files, log files etc). The metadata based patient profile with standalone and PIDNUM based format has been shown in Fig. 4.
4
Page 5 of 16
Metadata Wound Image
Patient’s Clinical Information
Upload Name Image 1
ID Sex
Wound Site Previous Visit History Parents History Current Medications
us
Ethnic Origin
cr
Address at time of Diagnosis
ip t
D.O.B Image 2
Telephone Number Year of Diagnosis
an
Comments
Timer: (data/time: acquired and saved successfully) Fig. 4: Metadata Format for Patient Profile Creation
M
C. Patient Profile/ Client end Approach
A chronic wound effected patient needs and wants a remote care, make contact with TMAs who are having networked camera based smart-phones; it creates his account in the apps and feeds the form with relevant clinical information. Every patient has his
d
profile once accounts create. Patients are asked to feed their name, wound site, wound images and also some comment relating to
te
their wound after which they are redirected to their profile and a PIDNUM is generated in the wound database which is unique for every patient. The PIDNUM has to keep in mind to the patient’s for future treatment, no need to carry hardcopy multiple
Ac ce p
prescriptions. Every detail of the patient is saved in the global wound database. Profile of patients shows their personal information including wound images and also link to profiles of doctors who have their account in TWM centre. This database stored registered patient’s detail and authorized best dermatologists information, so agents can help to show the details information to select the doctors. Before a patient selects a doctor for diagnosis of his wound he visits doctor’s profile and checks out information such as his qualification, experience and number of cases solved. It gives a clear picture to the patient whom to select for their diagnosis. Our proposed model introduces automatic query system based on wound type which automatically suggests the best doctor’s in their disease. There is a button on the bottom part of every doctor’s profile clicking on which the report of the patient is sent to the concerned doctor. The PIDNUM is saved in the database of that particular doctor. It takes some time for the patient to get e-prescribed medication and feedback from the doctor. Fig. 5 depicts the screenshot of patient centric data collection by TMA using Android operating system based Xolo Q1000 Opus smart-phone.
5
Page 6 of 16
cr
ip t Ac ce p
te
d
M
an
us
(a) Patient’s Clinical Information
(b) Patient’s wound Image
Fig. 5: Patient’s Data Collection by TMA through Mobile Interface
D. Doctor’s Profile/ Server end Approach
Expert doctors keep on checking their profile for reports from patients. When a doctor logs in the apps and is redirected to his profile he finds that he has a case which asks for care and diagnosis. He clicks on the link on the arrived case which says “visit”. This redirects doctor to a page ethic shows all the fed details of patients. Analysis of pictures of wound site and comments of patient makes clear the intensity of injury and seriousness of wound. Now the doctor has two options. Either he can go ahead and analyse the wound or can refer the report to other existing doctors on the apps. In case the report is referred, PIDNUM of that particular patient goes in the database of the doctor to whom the report is forwarded. If doctor chooses to diagnose the patient himself he goes in to analyse the report even more deeply by segmenting the wound site picture and sends the picture for colour correction in computer added diagnostic (CAD) tool. The image processing is also being there at server end in CAD tool. Once the wound site images are colour corrected and segmented based on the colour, texture and tissue type study of wound site reveals the type of wound patient has. Doctor now goes ahead and feeds in some prescription and feedback. Once the patient is treated and doctor makes it sure that the patient is treated he clicks on the “treated” button and it increases the cases solved by 6
Page 7 of 16
that particular doctor counter increment by 1. The report of the patient goes in the solved cases section of the doctor. At later stage whenever doctor finds it necessary to check out the cases solved by him he goes ahead and from the drop down menu on his profile which has two links “present cases” and solved cases. Solved cases link gives the report of the patient who are already treated by the doctor. Whenever a patient is treated the flag in doctor’s database is reset and it goes in solved cases list. Patients keep on checking their profile after they send their report to doctors. Doctors regularly keep sending prescriptions and feedbacks which are showed on the “doctor’s feedback” screen on patient’s profile. It helps patients who are in remote areas and do not
ip t
have access to expert care and advice. There is also a provision that patients can update their report by clicking on “update” button present on their profile. By updating their report patients can keep doctors updated with their status of wound and it can be brought about by either dropping in comments or uploading pictures of wound site. An update from patient is reflected in the
cr
database and a flag in doctor’s database is also set which reveals that report of the particular patient needs to be reviewed. This TWM apps helps people in remote sites of the world to have expert advice from doctors who are highly skilled and professional.
an extra burden. Fig. 6 shows the prototype implementation from doctor side.
an
E. Wound Database
us
It also helps elderly people to have home care in case of periodic checkups for whom transportation from remote site to clinic is
The database would be even stronger to provide the patient and doctor with even more dynamic features that allows more doctor-patient smooth interaction and vice-versa. If after diagnosis the doctor finds that site of wound has reached threshold
M
level, then will ask the patient to visit him or any other doctor nearby to the patient’s location which is judged from the address of the patient. The patient is also registered in the database of the visiting patient and given a date and time slot subject to availability. This is a favor which doctor will do on the patient. The date and time slot will be texted to the patient on his
d
telephone number and will also be updated on his profile. Medetec wound image database [20] has been used for this work. The medical image databases like Cumulative Index to Nursing and Allied Health Literature (CINAHL), Medline, nursing and
te
medical online journals, the British Nursing Index (BNI), The Cochrane Library, internet sources, and Department of Health
Ac ce p
(DoH) databases were searched for accessing the wound images.
(a)
7
Page 8 of 16
cr
ip t te
d
M
an
us
(b)
(c)
Ac ce p
Fig. 6 (a) (b) (c): Prototype Implementation with Patient’s ID and Feedback Panel
F. Metadata Protection
The major security issues will arise when transferring the metadata from smart-phone to TMH. Privacy and security both of are the important in any healthcare information systems. Patient’s clinical information should be always being considered as sensitive in general and identifying data should remain strictly confidential. Security mechanisms are required to protect the patient data, prevent improper system use and to insure the confidentiality. Each patient is assigned a unique PIDNUM and password for system sign-on and use. The mHealth [21] mainly use of mobile and wireless technology that enables doctors to remotely monitor patients health regularly, managing patient care, reducing drug shortages at health clinics, improve quality health care unit that’s why need to generate novel security and privacy issues [22], reducing cost and allowing patients to spend less time in the clinic. According to National Committee for Vital and Health Statistics (NCVHS), PHI as ‘personal health information’ rather than ‘protected health information’ which is an axiom that has specific meaning in a HIPAA perspective [23]. III. PROTOTYPE WOUND MONITORING SYSTEM This developed platform allows doctors to measure the progress of a wound timely. The paper makes an attempt to make a system that can efficient tool using PHP/MySqL simulation. Patient’s clinical information and wound images are linking with metadata and embedded in JPEG that is globally accessible through a web application. Metadata is also stored in the MySQL 8
Page 9 of 16
database. The number of times the patient visits his profile and updates something it is maintained in the database. And after every update the seen flag will be reset to “Not Visited” and that prompts the doctor to revisit the patient’s profile and rediagnose the report and send another feedback. The key use cases include: (a) Update patient clinical status, (b) Store wound images time-to-time to the database, (c) Update privacy code, (d) Delete arbitrary registry metadata and (e) Unchanged the PIDNUM. The Data required for this paper work was chronic wound patient information collected from a Medetec database. The chronic wound patient dataset contains around 100 records from a clinical computer system. The records are patient’s summaries
ip t
of various stages of wound images in the form of HTML pages. The portable devices like smart-phones, tablets, laptops with inbuilt camera capture wound images and computing and networking features allow smooth interaction. The MySqL/PHP based
cr
specific implementation staffs are presented below:
Step 1: Creation of patient’s demographic page using following commands
us
Step 2:
$fpass=$_POST['ide']; //PIDNUM is posted from form to the login page and is saved in
mysql_connect('localhost','root','')or die(mysql_error()); //mysql server is started mysql_select_db('medical')or die(mysql_error());
d
// concerned database is selected
M
$fpass variable
an
//beginning of php code
te
$result=mysql_query("SELECT PIDNUM FROM account")or die(mysql_error()); //id of all the patients are selected and saved in an array
Ac ce p
while($row=mysql_fetch_array($result)){ if($row['id']==$fpass){
// if PIDNUM of the current user matches with that present in array then the user is redirected to his own very profile else he/she is asked to create account $bunt=1;
$myfile="login.txt";
// retrieves the PIDNUM of the current patient $f=fopen($myfile,'w')or die("can't open file"); fwrite($f,$fpass); fclose($f); break; } else{ $bunt=0; } } ?>
9
Page 10 of 16
//end of php code Step 3: if ($success) redirect to patient’s profile else echo “create your account”
ip t
redirect to sign up page
cr
A. Flowchart:
The encoded data format should be sent from smart phone to TMH using Internet. The general person can get the doctors
us
details from TMA but not having permission to access the database or create a bulk doctor profile. The first step is SignUp as patient. Any person who wants to get expert advice from a doctor about his wound or wants his wound to get diagnosed creates
an
his account wherein he has to feed in some details along with details of wounded images area. The completion of these steps redirects him to his profile and from his profile page he can select a doctor for the diagnosis purpose. Another link on the page says sign in as patient. Someone who already has account on this application has unique PIDNUM and using that he tries to
M
login. If the login PIDNUM is valid he is redirected to profile page else it says not a registered user and he is asked to create account. On the profile page patient finds feedback and prescribed report from doctor in case his report is visited by doctors and seen. The third link on the application home page says sign in as doctor. This link is clicked by a doctor to visit his profile
d
wherein he finds reports from patients from various places and waiting to be visited and diagnosed. If the doctor finds case to be well and good enough to be handled by him he visits and goes ahead for diagnosis else he refers that particular case to other
Ac ce p
te
doctors catering to TWM application. Fig. 7 depicts the flow-map of TWM system.
Fig. 7: Tele-wound Monitoring Flow-map
10
Page 11 of 16
IV. RESULTS Every patient is asked to feed the following fields: clinical information like name, age, sex, date of birth, address and contact details etc and also taken wound images using smart-phones by TMA and send to TMH for best daigonosis. The submitted patient report stored in a global wound database with specified PIDNUM that serves as the key for retrieving the details from database. Fig. 8 has shown that the screenshot of doctor accessing the patient detail’s.
ip t
Patients Profile:
A. Login/Sign up – Patient’s information protection is big concern for maintaning confidentiality to enter a username and password.
cr
B.Chronic wound list – The location of wound effected portion is used to organize the wound list. The TMA easily see the
te
d
M
an
us
different types of wounds.
Fig. 8: Screenshot of wound lists in smart phone
PIDNUM.
Ac ce p
C. Patient history – The wound image and clinical information of the patient has been stored in the database with unique
D. Treatment – It allows a user to input information regarding any changes in wound images after getting prescribed medication by doctor. The doctor can visit patient’s images over a certain time interval. The previous treatment can be accesssed from patient history.
E. Wound History – The global community database has been developed to store patient’s information and as well as maintain the doctor’s profile.
The Fig. 9 depicts the screenshot of doctor’s accessing the patient details. The Fig. 10 below shows the screenshot of doctor’s seeing medical database including patient’s clinical information and wound images.
11
Page 12 of 16
ip t cr us
Ac ce p
te
d
M
an
Fig. 9: Screenshot of Doctor’s accessing the Patient Detail’s
Fig. 10: Screenshot of Doctor’s Profile
V. CONCLUSION AND FUTURE SCOPE This paper presents a web-enabled flexible informatics systems based framework for analyzing and processing wound images. The wound images have been taken from patient body through high resolution camera assisted networked mobile phones and stored in metadata (having efficient saving scheme) format and send it to CAD tool in TMH by the TMA remotely. The patients can get prescribed medication from the doctors with respect to unique PIDNUM. This approach gives a better response in terms of unique PIDNUM creation; patient can move one doctor to another having this unique PIDNUM, so full patient’s history easily 12
Page 13 of 16
accessible by several doctors. The CAD tool having image processing toolbox but it is not setup at client sides because of multiple number of agents are sending the patient’s details that embed image and clinical information to the global community database and cost is also highly demanding. It gives a clinically effectiveness, highly efficient, timely, more comfort and stress, reduced need for hospitalization and user friendly performances. The global wound database would be even stronger to provide the patient and doctor with even more dynamic features that allows more doctor-patient interaction vice-versa. In future we will propose an algorithm for Tele-truma case where patients can upload their images and videos of wound for better treatment. The
ip t
advanced metadata protection mechanisms are required for sharing wound images. The big challenges in managing large volumes of medical data that can be integrate with electronic health records for big data analytics purpose. This novel approach
cr
can be used in data mining application and public health monitoring systems. REFERENCES
us
[1]Gupta N, Gupta SK, Shukla VK and Singh SP (2004), “An Indian community based epidemiological study of wounds”, Journal of Wound Care, vol. 13, no. 8, pp. 323-325, Sept. 2004.
[2]Big changes coming to the wound care market. (2014). Available: http://seekingalpha.com/article/1757332-big-changes-coming-to-the-wound-care-marketin-2014.
an
[3]Dimitrios I. K and Fotini L. T, “Automated pressure ulcer lesion diagnosis for Telemedicine systems”, IEEE Engineering in Medicine and Biology Magazine, vol. 26, issue 5, pp. 18-22, Oct. 2007.
[4]Kass M, Witkin A,Terzopulos D.,”Snakes: Active contour models”, International Journal of Computer Vision, pp. 321-331, 1988.
M
[5]Hettiarachchi N. D. J, Mahindaratne R. B. H, Mendis G. D.C. and Nanayakkara H. T.,” Mobile based wound measurement”, IEEE Point-of-Care Healthcare Technologies (PHT), Bangalore, India, pp.298-301, Jan. 16-18, 2013.
[6]Orlando R.E. Pereira, Paulo A.C.S. Neves, Joel J.P.C. Rodrigues, “Mobile solution for three-tier biofeedback data acquisition and processing”, Proc. of IEEE
d
GlOBECOM, Global Telecom. Conf., New Orleans, LO, pp.1-5, Nov. 2008.
te
[7]Mawston N., “Camera phones outsell digital still cameras in H1 2003 and beyond”, Wireless devices strategies, Strategy Analytics, Newton Centre, MA, 2003.
Ac ce p
[8]Farmer A, Gibson O, Hayton P, et al., “A real-time, mobile phone-based telemedicine system to support young adults with type 1 diabetes”, Inform Primary Care, vol. 13, pp. 171–177, 2005.
[9]Vahatalo M.A., Virtamo H.E., Viikari J.S., Ronnemaa T., “Cellular phone transferred self blood glucose monitoring: prerequisites for positive outcome”, Practical Diabetes International, vol. 21, no. 5, pp. 192 – 194, 2004. [10] Bambang Parmanto, Gede Pramana, Daihua Xie Yu, Andrea D Fairman, Brad E Dicianno, Michael P McCue, “iMHere: A Novel mHealth system for supporting self-care in management of complex and chronic conditions”, Journal of Medical Internet Research Mhealth Uhealth, vol. 1, iss. 2, pp. 1-16, Nov., 2013.
[11] Gargi U., “Modelling and clustering of photo capture streams”, Proc. of Multimedia, ACM Press, pp.47-54, 2003. [12] Malcolm S, “Smartphone apps, Permissions and privacy concerns and next steps”, TEKDESK: A division of the community opportunity and innovation network, pp. 1-94, July, 2013. [13] Toyama, K., Logan, R., Roseway, A., Anandan, P.Geographic Location tags on digital images, Proceedings of the 11th ACM International Conference on Multimedia, ACM Press; pp.156-166. [14] Risto S., Eric H., Anita W. and Marc D., “Metadata creation system for mobile images”, ACM MobiSys’04 Proc. of the 2nd Int. Conf. on Mobile systems, applications, and services, Boston, USA, pp. 36-48, June, 2004.
13
Page 14 of 16
[15] Chinmay C., Bharat G. and Soumya K. G., “A review on telemedicine-based WBAN framework for patient monitoring”, Telemedicine and e-health, Marry Ann Liebert, Inc, vol. 19, no. 8, pp. 619-626, Aug., 2013. [16] Stephane Meystre, “The current state of Telemonitoring: A comment on the literature”, Telemedicine and e-Health, Mary Ann Liebert Inc., vol. 11, no. 1, pp. 63-69, Feb., 2005. [17] Chase, H. P., Pearson, J. A., Wightman, C., Roberts, M. D., Oderberg, A. D. & Garg, S. K., “Modem transmission of glucose values reduces the cost and need for clinic visits”, Diabetes Care, vol. 26, no. 5, pp. 1475–1479, May, 2003.
ip t
[18] Risto S., Mikko V., Juha P. and Hanno N., “Mobshare: Controlled and immediate sharing of mobile images”, MULTIMEDIA '04 Proc. of the 12th Annual ACM Int. Conf. on Multimedia, USA, pp.724-731, Oct., 2004.
cr
[19] Paul A. Harris, Robert Taylor, Robert Thielke, Jonathon Payne, Nathaniel Gonzalez and Jose G. Conde, “Research electronic data capture (REDCap) - A metadata-driven methodology and workflow process for providing translational research informatics support”, Journal of Biomedical Inform., Vol. 42, no.
us
2, pp. 377–381. doi:10.1016/j.jbi.2008.08.010, April, 2009. [20] Medetec wound database: Available: http://www.medetec.co.uk/files/medetec_image_databases.html.
[21] World health Organization, “mHealth: New horizons for health through mobile technologies: second global survey on eHealth”, Global Observatory for
an
eHealth Series, Vol. 3, pp. 1-112, 2011.
[22] Al Ameen M, Liu J, Kwak K, “Security and privacy issues in wireless sensor networks for healthcare applications”, Journal of Medical System, Springer,
M
vol. 36, no. 1, pp. 93-101, March, 2010.
[23] Choi Y.B., Capitan K.E., Krause J.S and Streeper M.M, “Challenges associated with privacy in healthcare industry, implementation of HIPAA and security
d
rules”, Journal of Medical System, vol. 30, no. 1, pp. 57-64, Feb. 2006.
te
Chinmay Chakraborty (M’09) was born in Sabang, West Midnapur, India in 1984. He received the B.Tech. degree in Electronics and Communication Engineering from West Bengal University of Technology
Ac ce p
(WBUT), India in 2006. From 2006 to 2007, he was with the Dept. of Industrial Engineering and Management, Indian Institute of Technology, Kharagpur (IIT Kharagpur), India as a Research Consultant. He received the M.S. degree in Telecommunication Engineering from G. S. Sanyal School of Telecommunications, IIT Kharagpur, India in 2010. He is worked at ICFAI Univ., Agartala, India as Senior Lecturer. Currently He is working at Birla Institute of Technology, Mesra, India as Assistant Professor. His research interests include Wireless Body Sensor Networks, Telemedicine, and Communication Networks.
Bharat Gupta received B.E and M.Tech degrees in Electronics Engineering from Govt. Engineering College of Ujjain, Madhya Pradesh, India and Birla Institute of Technology in 2000 and 2003 respectively. The Ph.D. degree he received in the area of Wireless Communication in Health care Monitoring from the Department of Microelectronics and Telecommunication at Univ. of Rome, Tor Vergata, Italy, in 2011. He has been teaching at the Birla Institute of Technology, Mesra, since August 2004 where he is presently an Associate Professor of the Electronics and Communication Engineering. His teaching and research interests include FM UWB Communication, Energy Efficient MAC for WBAN, Vital Sign Monitoring and WSN. He has published more than 20 technical papers in various journals and. He is a member of GTTI.
14
Page 15 of 16
Soumya K. Ghosh is Professor in the School of Information Technology, Indian Institute of Technology, Kharagpur (IIT Kharagpur), India. His primary areas of research include Geospatial Databases and Services, Cloud Computing and Security. Prior to IIT Kharagpur, he worked for Indian Space Research Organization in
ip t
area of remote sensing and geographic Information systems for natural resource management. He did his PhD and M.Tech in Computer Science, from Department of Computer Science & Engineering, IIT Kharagpur. He did his BE in Electronics and Communication Engineering from National Institute of Technology (formerly, Regional Engineering College),
Ac ce p
te
d
M
an
us
cr
Durgapur, India. He has published more than 100 technical papers in various journals and conferences.
15
Page 16 of 16