Docstoc

A Location-based Health Information Service

Document Sample
A Location-based Health Information Service Powered By Docstoc
					A Location-based Health
Information Service
Applying geographic filters to enable access to immediately relevant
information

Maged N Kamel Boulos, PhD




  Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Agenda
    •   Background
         –   Location Matters
         –   The Power of Where
         –   Location-specific Health Information
         –   Location-based Services Defined
         –   Knowing Where Users Are
              • Stopgap Solutions: Choose Your Location
              • IP Geo-targeting/ Geolocation Solutions
              • Related Technology: Hypertag
    •   Aim
    •   Objectives
    •   Example Scenario
    •   Important Issues and Concerns Related to the Proposed Service
         –   Different Device Capabilities
         –   User Privacy
         –   Resource Pool Visualisation and Navigation Issues
         –   Other Issues of Concern
    •   Research Potentials and Opportunities
    •   Conclusion
                   Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Location Matters - 1
      • The concept that location can influence health is
        well known in medicine and public health
        science.
      • Certain diseases tend to occur in some places
        and not others. Health information needs and
        services also vary with location.
      • Different places on Earth are usually associated
        with different profiles that can also change with
        time: physical, biological, environmental,
        economic, linguistic, social, cultural, and
        sometimes even spiritual profiles, that do affect
        and are affected by health, disease, and
        healthcare.
                Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Location Matters - 2
      • On the following two slides I am using GIDEON
        (Global Infectious Disease and Epidemiology
        Network—<http://www.gideononline.com/>) to
        diagnose a case of splenomegaly (no other clinical
        findings were input to the program).
      • The two screenshots show the ―Diagnosis Results‖
        (with probabilities) for the same patient but with a
        different country of disease acquisition in each
        screenshot (Kenya and United Kingdom in this
        example).
      • You can clearly see how the differential diagnosis and
        probabilities differ in each case by just changing the
        geographical location of disease acquisition.

                Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Location Matters - 3




            Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Location Matters - 4




            Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
The Power of Where - 1
     • Brute health information delivery risks
       overloading users with unnecessary
       information that does not answer their actual
       needs, and might even act as noise, masking
       any other useful and relevant information
       delivered with it.
     • Caregivers need to know not only the history
       of patients they treat but also information
       about the social and environmental context
       within which those patients live.


              Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
The Power of Where - 2
     • Patients and the public in general also have
       similar needs that vary with location.
     • A big challenge for online health information
       services today remains to find and push
       location-specific knowledge to users based
       on their location and associated needs
       (location awareness).

       Location matters, and the Internet should be
       no exception to this rule.



              Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Location-specific Health Information
      • Examples include:
         –   local disease rates, maps and guidelines;
         –   targeted health education;
         –   addresses of local healthcare facilities;
         –   local health news;
         –   local weather, pollen and air quality alerts and maps (e.g.,
             for asthmatics);
         –   local health risks and hazards;
         –   travellers’ health information;
         –   local drugs/ drug trade names and prices (in local
             currency);
         –   information whose digital distribution rights are limited to
             some location(s);
         –   in addition to serving up content (and interface) in
             language(s) relevant to the viewer’s location.

         Fact: 75% of the world’s population do not speak English.

                  Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Location-based Services Defined
     • Location-based services blend information about a
       person’s location with other useful content, providing
       relevant, timely and local information to consumers when
       and where they need it (IBM).
     • The market for location-based services is expected to
       reach US $20 billion by 2005, according to industry
       analyst Ovum (<http://www.ovum.com/>), especially with
       the advent of the wireless Web and the convergence of
       different information delivery media.
     • Location-based services need not be limited to mobile
       devices or to routing (e.g., where is/ how to reach the
       nearest pharmacy) and service dispatch functions. The
       field of application of location-based services is much
       wider than mobile devices only (Hogeweg, 2001).

                Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Choose Your Location - 1
     • The inability of online information services to
       deliver relevant content on the fly has forced the
       adoption of stopgap solutions like asking the user
       to ―choose location‖ (Parekh, 2002).
     • After Web site visitors have made the effort to
       choose their city, country and language, they try
       to find the information they are looking for. The
       effort is entirely made by the visitor, but on the
       Internet usually visitors do not want to make too
       much effort. If the effort required is too high,
       visitors will leave, and maybe go to see what
       competitors are offering (van Leeuwen, 2001).

              Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Choose Your Location - 2




           Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Choose Your Location - 3
     • Allowing the user to manually enter his/ her
       location (and other ―user profile‖ information if
       they wish so) remains a good option
       (complementary to automatic IP geolocation)
       and I am not excluding it, e.g., to offer visitors
       the possibility to choose a different language
       and/ or set of location-specific content other
       than that automatically chosen by the service.




              Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
IP Geo-targeting - 1
      • Internet Protocol (IP) geolocation is the science
        of determining the location (up to city and
        sometimes postal code levels) of a Web site
        visitor based on his/ her IP address (Quova—
        <http://www.quova.com/>).
         IP addresses consist of four blocks of numbers, e.g.,
         138.40.220.250. They can be compared to telephone
         numbers and are needed by computers on the Internet to
         communicate with each other. It is impossible to
         communicate without IP addresses. A maximum of 4.25
         billion IP addresses can be issued because the four
         blocks range from 0 to 255.



                Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
IP Geo-targeting - 2
      • IP geolocation does not use any DNS reverse
        look-ups, or WHOIS look-ups to determine a
        visitor’s location. They use very frequently
        updated proprietary databases to resolve a
        visitor's IP address to the corresponding
        geographical location.
      • IP Geo-targeting can be seen as the GPS (Global
        Positioning System) of the Internet. Internet
        services that make use of geo-targeting will travel
        with (and adapt to) users, wherever they go
        (Hogeweg, 2001).
      • IP targeting enables Web information services to
        recognise the geographical location of visitors in
        real-time and serve content relevant to location.
                Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
IP Geolocation Solutions - 1




     Source: Quova GeoPoint—<http://www.quova.com/>

                    Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
IP Geolocation Solutions - 2




     Source: Digital envoy NetAcuity—<http://www.digitalenvoy.net/>


                     Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
IP Geolocation Solutions - 3




                                                 Source: Geobytes GeoSelect—
                                                 <http://www.geoselect.com/>
             Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
IP Geolocation Solutions - 4
      • InfoSplit—<http://www.infosplit.com/>
      • GeoIP City Edition—<http://www.maxmind.com/app/city>
      • The National Security Agency: Network Geo-location
        Technology—<http://www.nsa.gov/programs/tech/factshts/20020506.htm>
         The National Security Agency, America’s cryptologic
         organisation, claims its methodology is very accurate, does not
         produce any false positive results and will scale well compared
         to other methods when the world moves from IP version 4 with
         32 bits of address space to IP version 6 with 128 bits.

          Related Articles:
          van Leeuwen A. Geo-targeting on IP Address - Pinpointing Geolocation of
          Internet Users. GeoInformatics. July/August 2001 -
          <http://www.geoinformatics.com/issueonline/issues/2001/07_2001/pdf_07_2001/28_31_iptar.pdf>
          Hogeweg M. Relocation Based Services. GeoInformatics. September 2001 -
          <http://www.geoinformatics.com/issueonline/issues/2001/09_2001/pdf_09_2001/13_hogeweg.pdf>




                       Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Hypertag Technology - 1

       • Location-based services need to know where
         users are.
       • Hypertag (<http://www.hypertag.co.uk>), a
         Cambridge-based company, has created
         cheap, smart tags that can be installed in
         information posters and adverts on the street to
         beam Web links to mobile phones and PDAs.
       • Consumers who see an interesting advert can
         point and click their phone or PDA at the advert
         to instantly access corresponding Web-based
         information.
       • Online content is thus delivered where it will be
         most useful and relevant.

                Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Hypertag Technology - 2
                                    Targeted health education example
                                                          (Idea: MN Kamel Boulos)




                                 Link to
                          http://www.cdc.gov/
                          ChooseYourCover/




              Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Aim
      • To develop a pilot location-aware online
        health information service targeting
        caregivers, patients and the public in
        general (with different content for different
        user roles).
      • This implies the development of a
        localised clinical/ health content server
        with the functionality to customise content
        to the location and needs of the viewer in
        real time.

               Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Objectives at a Glance

    Metadata*                       Reasoning                        Optimised User
                                  with Metadata**                     Experience
 Collect/ Describe                        Select                            Present

   Personal Profile
                                                                             Goal:
                                                                       For every user,
    Device Profile                 Content (Resource)
                                                                         always serve
                                       Selection
                                                                           the right
                                     and Formatting
  Location Profile                                                   personalised content
                                     Models/ Rules
                                                                       in suitable form
Resource Descriptions                                                     and format

                           Enabling access to information that is immediately relevant to users

           *Including spatial (where) and temporal (when) metadata about user and resources
                                          **User’s spatial history, if known, can improve results


                           Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Objectives - 1
      • To apply suitable methods for gathering and
        storing user profiles, including detecting their
        location (online user-filled forms and automatic
        location detection using for example GeoSelect IP
        geolocation technology–see
        <http://www.geoselect.com/Demo.htm>);
      • To determine the different location profiles of
        target users (a location profile describes the
        language, health and healthcare makeup/
        problems and corresponding clinical/ health
        information needs associated with that location);

         Metadata ontologies are needed to define and store user,
         device, location and resource characteristics and
         relationships (profiles or descriptions).

                 Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Objectives - 2
      • To develop the necessary content selection
        models or rules for different locations;
      • To create a suitable clinical/ health content
        (metadata) pool for the proposed service based
        on existing, freely accessible Web resources and
        health news sources (plus any suitable resources
        developed in-house). Selected resources must
        be adequately indexed regarding topic,
        provenance, coverage or scope, language,
        intended audience and other relevant aspects, in
        order for the proposed service to be able to
        unambiguously match content to location and
        user;
          ―If you do not index it, it does not exist. It is out there but you cannot find it,
                                                           so it might as well not be there.‖
                                     —Barbara Quint, ASI San Diego Conference, 1994
                   Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Objectives - 3
      • To implement a suitable language, interface and
        content customisation engine that can act on all
        of the above metadata and selection rules to
        always serve the right content in suitable form
        and format;
      • To continually evaluate the service during its
        development; and
      • To regularly document and publish results of the
        above steps in internal reports and appropriate
        peer-reviewed journals.
      • There is also a possibility of submitting the
        clinical/ health-specific metadata frameworks that
        are expected to arise out of this project to the
        appropriate standards bodies.
                Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Example Scenario - 1
     • IP targeting enables Web information
       services to recognise the geographical
       location of visitors in real-time, e.g., at the
       instant someone enters a Web site, it is
       recognised that the visitor is from Illinois,
       USA. Based on that knowledge, content can
       be shown which is likely to be relevant to
       visitors from this country, region or city.




              Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Example Scenario - 2
    Health content tailored to suit the needs of a visitor accessing the
    proposed IP-based/ location-based health information service from
    Illinois, USA, on 23 October 2002
    Visitor IP: 163.191.183.220
    Location detected: Springfield, Illinois, USA
    Location-specific Health Problem: West Nile Virus (WNV)




                    Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Example Scenario - 3
    Health content tailored to suit the needs of a visitor accessing the
    proposed IP-based/ location-based health information service from
    Illinois, USA, on 23 October 2002

    WNV Latest News




                    Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Example Scenario - 4
    Health content tailored to suit the needs of a visitor
    accessing the proposed IP-based/ location-based
    health information service from Illinois, USA, on
    23 October 2002

    WNV Prevention Information




                     Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Different Device Capabilities
      • User devices used to access a service might
        change with location, e.g., a desktop or laptop
        computer at home or in the clinic and a more
        limited mobile device on the road.
      • The drawback of the small size of mobile devices
        is that display is considerably smaller and input
        much more difficult (e.g., no full-scale keyboard).
      • Location-based services should ideally take into
        consideration the input and output characteristics
        of different devices by carefully choosing,
        personalising and formatting the content to
        display on such devices.
        Photo Caption: The first phone powered with Microsoft® SPV* Smartphone 2002 will
        be available at retail on 11 Nov 2002 in the UK for £180 (*SPV = Sound Pictures Video).

                     Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
User Privacy - 1
      • IP geolocation alone does not rely on cookies,
        profiles, registration data, or any other privacy-
        invasive techniques to identify the geographic
        location of an Internet user’s IP address and is
        incapable by itself of collecting other personally
        identifiable information.
      • Moreover, because IP addresses can be shared
        by hundreds of different devices (users) over the
        course of a few hours, individual privacy is further
        ensured (van Leeuwen, 2001).




                Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
User Privacy - 2
      • Consumers’ personal privacy becomes an issue
        of concern when extra personal information is
        collected besides IP location for enhanced health
        content personalisation (e.g., age, gender,
        occupation, etc., using user-filled forms) and
        when cookies are used to memorise preferences
        and track users.
      • Services should publish their Privacy Policy and
        respect consumers’ choices in this regard, or
        better still adopt the emerging P3P initiative
        (Platform for Privacy Preferences—
        <http://www.w3.org/P3P/>).

               Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
User Privacy - 3




      P3P is a machine-readable vocabulary and syntax for expressing a Web site’s data
      management practices. A site’s P3P policies present a snapshot summary of how
      the site collects, handles and uses personal information about its visitors. P3P-
      enabled Web browsers and other P3P applications will read and understand this
      snapshot information automatically, compare it to the Web user’s own set of privacy
      preferences, and inform the user when these preferences do not match the
      practices of the Web site he or she is visiting.
      Figure and Caption Source: <http://p3ptoolbox.org/guide/section2.shtml#Iia>

                         Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Resource Pool Visualisation and Navigation Issues

        • As online information portals accumulate metadata
          descriptions of Web resources, it becomes necessary
          to develop effective ways for visualising and
          navigating the resultant huge metadata repositories
          as well as the different semantic relationships and
          attributes of described Web resources. Interactive
          graphical maps provide one good method to visualise,
          understand and navigate a world that is too large and
          complex to be seen directly like the Web.
          Maps based on familiar metaphors taken from users’
          everyday life are much easier to understand.
          Associative and pictorial map icons that enable
          instant recognition and comprehension are preferred
          to geometric ones and are key to successful maps for
          browsing medical/ health Internet information
          resources.

                  Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Presentation
Matters
Recognition Not Recall
Screenshot of parts of
HealthCyberMap† and Visual
Net navigational maps for
resources on ―heart diseases‖.
Notice the difference in map
iconicity between
HealthCyberMap and Visual
Net approaches, and the map
clutter resulting from Visual
Net’s way of representing each
resource directly on the map
using a distinct point symbol.


             †The author’s PhD
                    project. Only
               HealthCyberMap
                 uses GIS and a
             clinical ontology to
                    classify Web
             resource data and
               render the maps. Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Other Issues of Concern - 1
     • The overall accuracy of IP geolocation varies
       between 95% and 98% depending on the
       currency, coverage, and granularity/ resolution* of
       the underlying geolocation provider database.
       This figure will never become 100% due to the
       existence of dynamic IP addresses and
       organisations that enter the Internet through one
       proxy server (van Leeuwen, 2001).
     • Users should be allowed to manually override IP
       locations determined by the service (if needed—
       this could be done once then stored in a personal
       profile that can be updated as often as
       necessary).
     •

         * Country only vs. country, region/ state, city and maybe also postal code.


                     Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Other Issues of Concern - 2
     • Other issues: information quality*, maintaining
       service currency, scalability of developed
       solutions, etc.




       * See: Kamel Boulos MN, Roudsari AV, Gordon C, Muir Gray JA. The Use
       of Quality Benchmarking in Assessing Web Resources for the
       Dermatology Virtual Branch Library of the National electronic Library
       for Health (NeLH). Journal of Medical Internet Research 2001;3(1):e5
       <URL: http://www.jmir.org/2001/1/e5/> [PubMed ID: 11720947]

                 Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Research Potentials and Opportunities - 1
       •   There are great potentials and opportunities for research into
           this innovative, cross-disciplinary topic that brings together the
           health informatics, health geographics, geoinformatics and
           Semantic Web communities.
       •   Location-based information means information that is
           immediately relevant, which is the essence of the Semantic
           Web (see <http://semanticweb.org>).
       •   Research literature on location-based health information
           services is currently very scarce (only one peer-reviewed
           paper by the author in PubMed/MEDLINE as of March 2003*).

       * Kamel Boulos MN. Location-based
       health information services: a new
       paradigm in personalised information
       delivery. International Journal of Health
       Geographics 2003 Jan;2:2 <URL:
       http://www.ij-healthgeographics.com/
       content/ pdf/ 1476-072X-2-2.pdf>
       [PubMed ID: 12556243]
                       Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Research Potentials and Opportunities - 2

       • More pioneering publications are expected to
         arise out of this research, besides its potential
         strategic value to international and national
         online health information services, for
         example:
          – National electronic Library for Health
            (<http://www.nelh.nhs.uk/>);
          – Public Health electronic Library/ Network
            (PHeL—<http://www.phel.gov.uk/>); and
          – NHS Direct Online
            (<http://www.nhsdirect.nhs.uk/>).



                Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Conclusion
     • Delivering real-time, location-enhanced and
       personalised health information and services can
       help consumers and providers accelerate and
       optimise their decision-making process in many
       medical/ health situations and problems.
     • The integration of a carefully selected variety of
       medical/ health Internet information services and
       resources with users’ tasks, needs, preferences
       and their device capabilities should enable users
       to focus more on informed decision-making and
       result in better health outcomes.


               Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>
Some Related Peer-reviewed Papers by the Author
       •   Kamel Boulos MN, Roudsari AV, Carson ER. Health Geomatics: An Enabling
           Suite of Technologies in Health and Healthcare (Methodolical Review).
           Journal of Biomedical Informatics 2001 Jun;34(3):195-219
           <doi:10.1006/jbin.2001.1015 - URL: http://www.idealibrary.com/ links/ doi/
           10.1006/ jbin.2001.1015> [PubMed ID: 11723701]
       •   Kamel Boulos MN, Roudsari AV, Carson ER. Towards a semantic medical
           Web: HealthCyberMap’s tool for building an RDF metadata base of health
           information resources based on the Qualified Dublin Core Metadata Set.
           Medical Science Monitor 2002 Jul;8(7):MT124-36 <URL:
           http://www.medscimonit.com/ pub/ vol_8/ no_7/2615.pdf> [PubMed ID:
           12118210]
       •   Kamel Boulos MN, Roudsari AV, Carson ER. A Dynamic Problem to
           Knowledge Linking Semantic Web Service Based on Clinical Codes.
           Medical Informatics & The Internet in Medicine 2002 Sep;27(3):127-137
           [PubMed ID: 12507259]
       •   Kamel Boulos MN, Roudsari AV, Carson ER. HealthCyberMap: A Semantic
           Visual Browser of Medical Internet Resources Based on Clinical Codes
           and the Human Body Metaphor. Health Information and Libraries Journal
           2002 Dec;19(4):189-200 [PubMed ID: 12485148]
       •   Kamel Boulos MN. The use of interactive graphical maps for browsing
           medical/ health Internet information resources. International Journal of
           Health Geographics 2003 Jan;2:1 <URL: http://www.ij-healthgeographics.com/
           content/ pdf/ 1476-072X-2-1.pdf> [PubMed ID: 12556244]

                      Concepts from HealthCyberMap—<http://healthcybermap.semanticweb.org>

				
DOCUMENT INFO
Shared By:
Categories:
Stats:
views:149
posted:3/25/2008
language:English
pages:42